A S U R V E Y OF CONVENTIONAL A N D N O V E L PROCESSES FOR THE T R E A T M E N T OF R E F R A C T O R Y G O L D by L A N C E M I C H A E L F A I R L E Y B.Sc , The University of British Columbia, 1976 A THESIS SUBMITTED IN PARTIAL F U L F I L L M E N T OF THE REQUIREMENTS FOR THE DEGREE OF M A S T E R OF APPLIED SCIENCE in THE F A C U L T Y OF G R A D U A T E STUDIES THE DEPARTMENT OF M E T A L S A N D MATERIALS ENGINEERING We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH C O L U M B I A August 1998 © Lance Michael Fairley, 1998 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, 1 agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of H*LWIS> * HqWicJU E w ^ ^ f t i L W ^ The University of British Columbia Vancouver, Canada Date 31 7-DE-6 (2/88) A B S T R A C T This thesis is a survey of treatments for the recovery of gold from refractory ores. An elucidation of the characteristics leading to gold retention and a definition for refractory gold ore is presented. From this survey of treatments are derived diagnostic conditions that may be useful in predicting the specific refractory characteristics of many ores. Conventional processes are surveyed and include the refractory ore treatments that enjoy commercial application. Oxidative pretreatments, lixiviants for gold dissolution, and adsorption methods are discussed for both conventional and emerging or less known refractory gold ore treatments. These emerging methods may have seen considerable research and possibly some commercial application. Included are some alternate techniques that were developed for specific ores and that may have been superseded by less expensive or more general treatments. Some diagnostic test conditions are proposed as being useful in elucidating the refractory components of non-cyanidable ores. Many pretreatments and lixiviating leaches are applied to sulfidic ores with cyanidation gold recoveries as low as 2.6% leading to extractions as high as 99%. Samples of moderately refractory carbonaceous ore with varying organic carbon and sulfide content are treated similarly and exhibit extraction gains from 66.5% to 92.3% T A B L E OF C O N T E N T S A B S T R A C T ii T A B L E OF CONTENTS iii LIST OF T A B L E S . v LIST OF FIGURES vii A C K N O W L E D G E M E N T viii 1.0 INTRODUCTION 1 2.0 LITERATURE REVIEW - THE DEFINITION OF R E F R A C T O R Y G O L D ORE 4 2.1 Cyanicides 4 2.1.1 Cyanide Consumption 5 2.1.2 Cyanide Degradation : 5 2.1.3 Oxygen Consumption 6 2.2 Occluded Gold 7 2.2.1 Tellurides 8 2.2.2 Antimonial Ores 8 2.3 Preg-robbing Ores 8 2.3.1 Carbonaceous Species 9 2.3.2 Clay Minerals 9 3.0 LITERATURE REVIEW - CONVENTIONAL R E F R A C T O R Y GOLD ORE TREATMENTS 11 3.1 Oxidizing Pretreatments . .11 3.1.1 Pre-aeration 12 3.1.2 Pressure Oxidation 13 3.1.3 The Redox Process 15 3.1.4 The Nitrox Process 19 3.1.5 Chlorine/Hypochlorite Pretreatment 19 3.1.6 Bioleach Oxidation 20 i i i 3.2 Lixiviation 23 3.2.1 Cyanide Lixiviation 24 3.2.2 Pressure Cyanidation 24 3.2.3 Thiourea Lixiviation 25 3.2.4 Chlorine 26 3.2.5 Bromine 27 3.2.6 Iodine 27 3.2.7 Other Lixiviants 28 3.3 Adsorption 33 3.3.1 Resin-in-leach 33 3.3.2 Carbon-in-Chlorine Leach 34 4.0 SCOPE A N D OBJECTIVES 35 5.0 R E S E A R C H RESULTS A N D DISCUSSION 36 5.1 Feed Characterization 36 5.1.1 Sample Identification 36 5.1.2 Head Sample Analysis 38 5.1.3 Mineralogy 41 5.2 Pretreatment 42 5.2.1 Experimental Design 42 5.2.2 Grinding 45 5.2.3 Diagnostic Lixiviation 46 5.2.4 Bioleach Oxidation 48 5.2.5 Thermal Pretreatments 5 2 5.2.6 Nitric Acid Leach 56 5.2.7 Sulfoxy Species from Sulfuric Acid, Peroxide and Persulfate 60 5.2.8 Oxidation with Hypochlorite in Brine 67 5.3 Lixiviation 71 5.3.1 Thiosulphate 7 1 5.3.2 Halides • • 74 6.0 CONCLUSIONS 77 BIBLIOGRAPHY 80 APPENDIX 90 i v LIST OF TABLES Table 1. Cyanide Species Dissociations 5 Table 2. Principal Gold Bearing Tellurides 8 Table 3. Gold Lixiviants 29 Table 4. Refractory Ore Head Analysis 38 Table 5. Elemental Ore Analysis 39 Table 6. Derived Mineralogy 40 Table 7. Grind Data 45 Table 8. Cyanidation of Ground Ore 46 Table 9. Ore Cyanidation Reagent Consumption 47 Table 10. Shakeflask Bioleach Parameters 49 Table 11. Stirred Tank Bioleach Parameters 50 Table 12. Bioleach Results 51 Table 13. Roasting Conditions and Results 55 Table 14. High Temperature Redox Results 57 Table 15. Low Temperature Redox Results - Sulfidic Ore 57 Table 16. Low Temperature Redox Results - Carbonaceous Ore 58 Table 17. Peroxide Oxidation 66 Table 19. Chlorination of Carbonaceous Ore With Cyanuric Acid Added 69 Table 20. Chlorination of Sulfidic Ore With Cyanuric Acid Added 70 Table 21. Thiosulfate Lixiviation 73 v Table 22. Mixed Halide Gold Extraction Table 23. Bromine/Bromide Gold Extraction LIST OF F I G U R E S Figure 1. Potential-pH equilibrium diagram for the gold-chloride system for [Au] = 10"5 M and [C1] = 10- 2 M 30 Figure 2. Potential-pH equilibrium diagram for the gold-bromide system for [Au] = 10"5 M and [Br] = 10" 2M 31 Figure 3. Potential-pH equilibrium diagram for the gold-iodide system for [Au] = 10"5 M and [I] = 1 0 ' 2 M 32 Figure 4. Experimental Plan 44 Figure 5. Carbonaceous ore stirred tank bioleach 51 Figure 6. Carbonaceous ore showing evidence of process activation 59 Figure 7. (a) Relationship between atomic orbitals, molecular orbitals, and the band molecular orbital diagram for the FeS 6 cluster, (b) Band model for FeS 2 (from Osseo-Asare) 62 Figure 8. CIL gold recoveries for 2844 ore after peroxide leaching 67 Figure 9. Gold extraction at 12 kg/t Br 2 and 6 kg/t KBr 75 v i i A C K N O W L E D G E M E N T I would like to express my sincere gratitude to Dr. David Dreisinger for his inexhaustible patience and guidance during the course of this study. I would like to thank Ms. Anita Lam for her assistance on many test campaigns and particularly for her mineral analysis skills and pleasant demeanor in the face of long days of sampling. Thanks also to Wonder Chigwida for a summer of technical assistance and Pinson Mining Company and the Cordex Exploration Company for the samples of ore. I am very grateful to Rayrock Yellowknife Resources Inc. and the Department of Metals and Materials Engineering at the University of British Columbia for their financial support. I am most grateful for the patience of my wife Maureen and the arrival of my new daughter Katherine. They both inspired the completion of this work. v i i i 1.0 I N T R O D U C T I O N The intent of this thesis is to examine past and current hydrometallurgical treatments applicable to the recovery of gold from non-cyanidable deposits. A definition for refractory gold is developed and the factors that contribute to refractoriness are described. The background literature search surveys what is known about the causes of gold retention by ores and reports the state-of-the-art of refractory gold ore treatments. In 1986, Dry and Coetzee concluded "at present the choice of a particular process route for the recovery of gold from any given refractory ore must be based on the results of empirical testing". This study summarizes conventional processes including the refractory ore treatments that enjoy commercial application. Oxidative pretreatments, lixiviants for gold dissolution, and adsorption methods are discussed for both conventional and emerging or less known refractory gold ore treatments. The emerging methods discussed may have seen considerable research and possibly some commercial application. Included are some alternate techniques that were developed historically for specific ores and that may have been superseded by less expensive or more general treatments. Treatments of refractory gold ores can be categorized for the particular refractory property that they are designed to address. A further distinction can be made between treatments suitable for low grade ore and those more feasible for high grade or beneficiated ores and concentrates. The focus of this study is to examine whole ore treatment methods that may be generalized for diagnostic purposes. 1 Some diagnostic test conditions are proposed as being useful in distinguishing between the refractory components of non-cyanidable ores. This combined with mineralogical assessment may reduce the empirical testing requirements referred to by Dry and Coetzee. A number of surveys of refractory gold ore treatments have been published in the last 30 years by various metallurgical institutes, engineering companies, groups of researchers and individuals. From the earliest survey reviewed (Nagy et al. 1966) to the most recent (Goode 1993) the consensus as to what constitutes a refractory gold ore appears to be any gold occurrence that cannot be substantially recovered by conventional cyanide leaching after grinding to a reasonable size (Mrkusic 1970). The cyanide leaching of gold has been an important hydrometallurgical process since its introduction in a patent by John MacArthur and Robert and William Forrest more than 100 years ago. An arbitrary level of extraction of 80% has been suggested as the division between free-milling and refractory gold ore (Fraser et al. 1991). The interest in the recovery of gold from refractory ores is a function of prevailing gold price and, more recently, the growing inventory of refractory gold reserves coupled with the exhaustion of easily recoverable gold. A large resource of refractory gold tailings from conventional extraction circuits has provided an easily accessible feedstock for metallurgical test work in the past, but further process development is likely to be incorporated in the exploratory and feasibility study costs of new mine prospects. As new mineral occurrences are discovered it is often found that new problems in the gold extractive process are brought to light (Scheiner et al. 1972). It is necessary to consider the mineralogical characteristics, processing options and economic factors when designing suitable 2 treatments for specific refractory gold ores. Some preliminary attempts at developing a knowledge base for an expert system have been published (Foo et al., Pendreigh & Barratt, Henley, Major et al.) but a detailed and exhaustive examination of ore mineralogy and the chemistry of process altered residues, for use as a predictive model for gold extraction from refractory ore, remains to be done. With the exception of the few strictly gravity concentration - direct smelting methods to produce dore, all gold recovery processes incorporate a hydrometallurgical step. The majority of the world's gold production requires the use of cyanide lixiviation and there exist a number of leaching or dissolution pretreatments as well as hydrometallurgical purification and stripping techniques that are applied to both free-milling and refractory gold ores. This study will focus primarily on hydrometallurgical processes with some cosideration of the pyrometallurgy of whole ore roasting. 3 2.0 L I T E R A T U R E R E V I E W - T H E DEFINITION OF R E F R A C T O R Y G O L D O R E There are many reasons for which an ore exhibits refractory behaviour. Often it is due to the nature and distribution of the gold itself, but the nature of the overall ore mineralization must be considered. It is often necessary to have a complete elemental as well as mineral analysis of an ore before the refractory nature can be understood. The major groupings of refractory behaviour are those where reactive minerals interfere with the cyanidation process, where physical locking of the gold within cyanide impermeable minerals prevents liberation and a special case where the cyanide or otherwise complexed gold is adsorbed onto a gangue component of the ore. Diagnostic leaching of specific minerals and their stability phases can often clarify the deportment of gold in an ore (Tumilty et al., Lorenzen & van Deventer). 2.1 Cyanicides Cyanicides are those constituents of an ore that reduce the activity of cyanide for the purpose of gold dissolution. They are represented by three main phenomena and are characterized by the presence of reactive minerals in the ore that lead to cyanide consumption, irreversible cyanide degradation or the related case where reactive minerals cause excessive oxygen consumption interfering with the cyanidation process. 4 2.1.1 Cyanide Consumption Metal cations, particularly copper (Stanley), can compete for the cyanide ligand, forming stable complexes and reducing the cyanide activity in solution (Hefter and May). The stabilities of some cyanide complexes are indicated in Table 1 (Hsu). The major loss of uncomplexed cyanide remains volatilization, particularly if the pulp pH approaches 9.4 (Stoychevski & Williams). Table 1. Cyanide Species Dissociations Cyanide Complex Equilibrium Reaction K a o r K d H C N , CN" CN" + H 2 0 = H C N + OH" 2xl0- 1 0 Zn(CN) 4 2" Zn(CN) 4 2" - Zn 2 + + 4CN" io-16 Cd(CN)3" Cd(CN)3" = C d 2 + + 3CN" io-19 Ni(CN) 4 2" Ni(CN) 4 2- = N i 2 + + 4CN- io-21 Cu(CN) 3 2" Cu(CN) 3 2-= Cu + + 3CN" io-29 Fe(CN)6 4" Fe(CN)6 4" = Fe 2 + + 6CN" IO"35 Ag(CN) 2" Ag(CN) 2" = A g + + 2CN" io-41 Fe(CN)6 3" Fe(CN)6 3" = Fe 3 + + 6CN" IO"42 2.1.2 Cyanide Degradation At potentials that can exist in cyanidation of auriferous sulfides it is possible to produce both cyanate and thiocyanate effectively reducing gold dissolution capacity. 5 2S 2 0 3 2 - + 0 2 + 2CN" = 2SCN" + 2S0 4 2" 0) 2FeS + 0 2 + 2H 2 0 + 14CN" = 2Fe(CN)64" + 2SCN" + 40H" (2) Fe 5S 6 + CN" = SCN" + 5FeS (3) CN" + S 0 2 + 0 2 + H 2 0 - CNO" + H 2 S 0 4 (4) Copper is known to catalyze the oxidation of cyanide to cyanate although the reaction is slow at the pH normally encountered in cyanidation. This reaction in the presence of sulphur dioxide and a copper catalyst is the Inco S0 2 -Air cyanide destruction process which is very effective at a potential of 110 m V S C E at a pH of 10 (Devuyst et al.1982), equation (4) being essentially irreversible. 2.1.3 Oxygen Consumption Eisner [1846] recognized that oxygen was essential to the dissolution of gold by cyanide solutions. Bodlander [1896] suggested that dissolution should proceed through a two step process with the formation of a hydrogen peroxide intermediate (Habashi): 2Au + 4NaCN + 0 2 + 2H 2 0 - 2NaAu(CN) 2 + 2NaOH + H 2 0 2 (5) 2Au + 4NaCN + H 2 0 2 = 2NaAu(CN) 2 + 2NaOH (6) 4Au + 8NaCN + 0 2 + 2H 2 0 = 4NaAu(CN) 2 + 4NaOH (7) Minerals which decompose in the alkaline cyanidation solutions to form polysulfide anions, sulfoxy anions (polythionate, thiosulfate), thioarsenite, thioantimonite and various other oxidizable species 6 tend to compete for the dissolved oxygen necessary for gold dissolution (Nagy et al., Komnitsas and Pooley 1989). Pyrrhotite is particularly reactive as an oxygen consumer (Graham). 2.2 Occluded Gold Siliceous refractory ores can lock gold into the gangue phase. This phase can be made up of microcrystalline quartz, chalcedony, chert, or amorphous forms of opaline silica. The gold can be present either as microfine metallic gold or auriferous sulfides occluded in the impermeable matrix (Hausen 1989). This type of gold occurrence can lead to size dependant refractoriness where gold cannot be liberated with a reasonable input of grinding energy. An important variation of this is where the gold exists as a microfine dispersion or even a lattice substitution in a sulfide mineral. Submicron gold has been found in pyrite, arsenopyrite, pyrrhotite and chalcopyrite in levels up to 20 ppm (Chryssoulis et al. 1988) and 488 ppm in pyrite (Swash). Arsenic rich zones within a mixed sulfide matrix have been shown to contain as much as 6900 ppm gold (Claassen). To release this gold it is necessary to alter the matrix, generally by a destructive oxidation, since grinding to submicron levels is impractical. Auriferous sulfides represent the majority of refractory gold ores. Other refractory gold ores contain gold solid solutions that have been found to resist dissolution by cyanide. These commonly contain gold compounds with tellurium, antimony, arsenic, selenium and occasionally lead or bismuth (Vaughan, Goode). 7 2.2.1 Tellurides Solid solutions of gold in tellurium are resistant to extraction with cyanide. Table 2. Principal Gold Bearing Tellurides Calaverite AuTe 2 Sylvanite ' AuAgTe 2 Montbrayite Au 2 Te 3 Petzite Ag 3 AuTe 2 Krennerite (AuAg)Te 2 Kostovite AuCuTe 4 2.2.2 Antimonial Ores Antimonial ores generally contain stibnite occluded gold and occasionally aurostibnite where the gold substitutes for antimony in the sulfide matrix. The dissolution of stibnite (Sb2S3) can lead to the formation of thioantimonites at high pH and it has been suggested that these can passivate a gold surface by adsorption (Stanley). 2.3 Preg-robbing Ores Gold ores containing various forms of graphitic or organic carbon can actively adsorb the dissolved gold cyanide species. This leads to a portion of the gold remaining bound to carbonaceous particles and reporting to the tailings. The term preg-robbing was first used by the Carlin Gold Mining Company to describe highly active types of carbonaceous ores. The mere presence of carbon is not 8 a reliable indicator of preg-robbing capacity since it is necessary for the carbonaceous gold ore to be an active gold adsorber. The ore can be inherently active or become activated by thermal or oxidative treatments. 2.3.1 Carbonaceous Species A variety of functional groups and active sites are available on active organic carbon including both oxidizing and reducing sites and many stereochemically specific surfaces. These sites have been partially described by infrared analysis to include most known simple aliphatic and aromatic combinations of carbon, oxygen and hydrogen with the presence of thiosulfate and amino functional groups also known. Organic (humic) acids are known to react with aurocyanide species preventing the complete recovery of gold from the ore (Afenya, Guay 1981). Graphitic or crystalline carbon may provide the primary adsorption sites for aurocyanide and may cause a polymerization to Au 4 (CN) 5 _ on the graphitic plate faces at acidic pH levels (Ibrado and Fuerstenau). 2.3.2 Clay Minerals Some clay minerals such as limonite and halloyite have been observed to provide adsorption sites for the gold cyanide complex. A study of the K 2 0 content present as the mineral illite ((K,H30)(Al,Mg,Fe)2(Si,Al)40,o((OH)2H20)) at one mine suggests that there is a correlation between 9 the clay content and the tendency of the carbonaceous ore to be "geo-activated" and become preg-robbing (Fuchs & Schurer). It is also possible that cationic sites for the adsorption of gold cyanide complex can be created by the initial adsorption of a cation from the dissolution of associated minerals. 10 3.0 L I T E R A T U R E R E V I E W - C O N V E N T I O N A L R E F R A C T O R Y G O L D O R E T R E A T M E N T S The various reasons for which gold ores exhibit refractory behaviour have spawned a number of approaches to the problem. It is often found that a combination of the reasons for gold refractoriness are found in a single ore. When both refractory sulfidic minerals and active carbonaceous components exist the ore is referred to as doubly refractory. Once the nature of an ore's refractoriness has been determined an appropriate treatment method can be chosen from the variety that are available. The difficulty that often arises is the fact that most methods were originally developed as site specific remedies and may not translate well to a new ore type. Some generally successful hydrometallurgical treatments have found commercial application at multiple sites and can be considered conventional. 3.1 Oxidizing Pretreatments Conventional methods for the oxidation of refractory gold ores and concentrates include both pyrometallurgical (roasting) and hydrometallurgical (pre-aeration, pressure oxidation, chlorine/hypochlorite oxidation, and bioleaching) pretreatments. 11 3.1.1 Pre-aeration The pre-aeration of refractory gold ores is generally used as a method to oxidize cyanicides before cyanidation. It includes any method that improves oxygen contact with unstable minerals at ambient pressure by sparged air or well vortexed agitation. It can be used in the pretreatment of slurries or heaps (Bartlett) and is most often done in the presence of a caustic reagent such as lime or sodium hydroxide. There is evidence of the leaching of unstable pyritic minerals such as the pyrrhotites, marcasite and framboidal pyrite at high pH in a pre-aeration stage. 4FeS + 50 2 + 4H 2 0 + 40H" = 4Fe(OH)3 + 2S 20 3 2" (8) 4FeS2 + 150 2 + 160H" = 4Fe(OH)3 + 8S0 4 2" + 2H 2 0 (9) S 20 3 2" + 20 2 + 20H- = 2S0 4 2" + H 2 0 (10) At the Golden Sunlight Mine, research has suggested that the pre-aeration of sulfides prevents the formation of thiocyanates and improves the free cyanide concentration in the gold dissolution stage (Lofftus et al.). 3.1.1.1 Cyanidation Promoters It is common that the pre-aeration treatment also incorporates the addition of cyanidation promoters. These consist of soluble lead or mercury salts which remove sulfide from solution before it begins to consume oxygen. It has been concluded that lead in solution can activate passivated sulfur 12 complexed gold surfaces, prevent gold sulfide film formation and act as a sulfide oxidation intermediary through lead sulfide formation (Kondos et al.) 3.1.2 Pressure Oxidation The extension of pre-aeration to higher temperatures and overpressures of oxygen is becoming a popular method for the pretreatment of refractory gold ores. Although the process in alkaline conditions is used commercially at the Mercur Mine, which has a high carbonate content ore, it is the acid process that is finding application around the world. An improvement in recovery is brought about by the oxidation of cyanicides and the destruction of the sulfide matrix which often occludes finely disseminated gold. 4FeS2 + 140 2 + 4H 2 0 = 4FeS0 4 + 4H 2 S0 4 (11) 4FeS0 4 + 0 2 + 2H 2 S0 4 = 2Fe 2(S0 4) 3 + 2H 2 0 (12) 4FeS2 + 150 2 + 2H 2 0 = 2Fe 2(S0 4) 3 + 2H 2 S0 4 (13) FeS 2 + 20 2 = FeS0 4 + S° (14) The elemental sulfur product is considered an intermediate (McKay & Halpern) and in high temperature pressure oxidation (>160°C), is itself oxidized. 13 2S° + 30 2 + 2H 2 0 = 2H 2 S0 4 (15) Considerable research has been done to clarify the mechanism of sulfide oxidation and it is possible that elemental sulfur is a low potential reduction product from a true intermediate such as thiosulfate or a polysulfide (Mishra & Osseo-Asare). Further hydrolysis reactions of the ferric sulfate produce more acid, requiring a neutralization step before cyanidation can be carried out. This requirement for neutralization often represents a significant cost factor when a ready source of limestone is not close at hand. Fe 2(S0 4) 3 + 2H 2 0 = 2Fe(OH)S0 4 + H 2 S 0 4 (16) Fe 2(S0 4) 3 + 3H 2 0 = 2Fe 2 0 3 + 3H 2 S0 4 (17) 3Fe 2(S0 4) 3 + 14H 20 = 2H 3OFe 3(S0 4) 2(OH) 6 + 5H 2 S0 4 (18) A hydronium jarosite product can incorporate various metals from the reaction slurry. This can be the cause of poor silver recoveries in pressure oxidation processes since silver jarosite is relatively insoluble in the gold extraction stage (Berezowski & Weir 1983,1989). Arsenopyrite exhibits some elemental sulfur production (Equation (23)) but the desired high temperature product in a fully oxidized system is ferric arsenate and sulfate (Papangelakis & Demopoulos 1990). 14 2FeAsS + 70 2 + 6H 2 0 = 2FeAsCy2H 2 0 + 2H 2 S0 4 (19) 4Fe 7S 8 + 690 2 + 10H 2SO 4 - 14Fe 2(S0 4) 3 + 10H2O (20) 4FeS + 90 2 + 2H 2 S0 4 = 2Fe 2(S0 4) 3 + 2H 2 0 (21) The pyrrhotite reactions approximated in equations (20) and (21) demonstrate a net acid consumption. A l l of these sulfides provide significant heat upon oxidation with arsenopyrite providing the most energy per mole of sulfide sulfur (Mason). A combination of ultra-fine grinding of sulfide flotation concentrates followed by low intensity pressure oxidation has shown some success in refractory gold ore treatment and is marketed as the Activox™ process (Johnson et al.). 3.1.3 The Redox Process The Redox process uses a nitric acid catalyzed oxygen leach to decompose refractory pyrite and arsenopyrite gold ores. The residue produced can be treated for gold extraction by conventional means such as cyanidation. It is because of the nitric acid catalysis that the leach is very rapid and complete. Variants of the process have been developed to optimize the treatment of materials with a range of pyrite and arsenopyrite contents (Beattie et al.). 15 3.1.3.1 Leach Chemistry Pyrite (FeS2) and arsenopyrite (FeAsS) are decomposed by nitric acid according to the following equations: The stoichiometric consumption of nitric acid by pyrite is 2.63 kg/kg and sulfate is the only product of sulfur oxidation. The stoichiometric consumption of nitric acid by arsenopyrite is 1.65 kg/kg and the reaction has approximately a 20% yield of elemental sulfur. Less stable minerals such as marcasite develop lower potentials at the surface and can decompose as follows: FeS 2 + 5 H N 0 3 = 1/2 H 2 S 0 4 + 1/2 Fe 2(S0 4) 3 + 5 NO + 2 H 2 0 (22) FeAsS + 4.27 H N 0 3 + 1.4 H + - Fe 3 + + 0.8 S0 4 2 " + 0.2S + H 3 A s 0 4 + 4.27 NO + 4/3H 20 (23) FeS 2 + 4 H N 0 3 = Fe(N0 3) 3 + 2S + NO + 2H 2 0 (24) This produces elemental sulfur which is difficult to oxidize further at low potentials. 16 The product of nitric acid reduction is nitric oxide gas, NO. Nitric oxide reacts with oxygen to form nitrogen dioxide, N 0 2 , which is absorbed as nitric acid. These reactions are shown in the equations below: 6 NO + 3 0 2 = 6 N 0 2 (25) 6 N 0 2 + 2 H 2 0 .= 4 H N 0 3 + 2 N O (26) 4 N O + 3 0 2 + 2 H 2 0 = 4 H N 0 3 (27) The regeneration of nitric oxide to nitric acid can be carried out concurrent with the leach or as a separate unit operation. Overall, the reactions of pyrite and arsenopyrite with oxygen in the Redox process, are shown in the equations (28) and (29): FeS 2 + 15/4 0 2 + 1/2 H 2 0 = 1/2 H 2 S 0 4 + 1/2 Fe 2(S0 4) 3 (28) FeAsS + 3.2 0 2 + 0.8 H 2 0 + 1.4 FT = Fe 3 + + 0.8 S0 4 2 " + 0.2 S°+H 3 As0 4 (29) By these equations the stoichiometric consumption of oxygen is 0.62 kg/kg arsenopyrite and 1.00 kg/kg pyrite. The initial reaction with arsenopyrite consumes one equivalent of acid per mole of sulfide reacted; pyrite generates one equivalent of acid. 17 In the atmospheric pressure leach used for low grade sulfide ores the nitric oxidant is regenerated outside of the leach vessel so that the final stage of a continuous circuit is nitrate depleted. The high pressure autoclave process uses the in-situ regeneration of nitric acid to maintain the reaction of oxygen with a sulfide concentrate. 3.1.3.2 Precipitation Reactions The Redox leach solubilizes iron, arsenic and sulfur, however, precipitation reactions can take place concurrently. These are favoured by high temperature and low solution acidity. The compounds precipitated include ferric arsenate, FeAs0 4-2H 2S0 4; basic ferric sulfate, Fe(SO 4XOH); and hematite, Fe 20 3 . Silver can be precipitated as refractory silver jarosite, AgFe 3(S0 4) 2(OH) 6 resulting in metal value losses to the tails. Fe 3 + + H 3 A s 0 4 + 2 H 2 0 = FeAs0 4 -2H 2 0 + 3 FT (30) Fe 3 + + 3/2 H 2 0 = 1/2 Fe 2 0 3 + 3 FT (31) Both reactions produce three equivalents of acid. The bulk of the sulfate must be removed from solution as gypsum or anhydrite. The calcium sulfate is produced by the addition of limestone to the leach slurry as shown in equation (32): C a C 0 3 + S0 4 2 - + 2 FT = CaS0 4 + C 0 2 + H 2 0 . (32) 18 It was observed in pilot plant trials of the process at Bacon, Donaldson & Associates' test facility in 1988 (unpublished) that the precipitation of sulfate promotes the precipitation of ferric arsenate probably by removing sulfate ligands that compete for the necessary ferric iron in solution although the pH increase to 1.5 due to concurrent acid neutralization also plays a part. The overall reaction for the oxidation of pyrite with the production of hematite proceeds as follows: FeS 2 + 2CaC0 3 + 15/402= l /2Fe 2 0 3 + 2CaS0 4 + 2 C 0 2 (33) In this reaction there is no overall production or consumption of acid. 3.1.4 The Nitrox Process This nitrate catalyzed oxidation method has been generally limited to the atmospheric pressure leaching of refractory gold ore using air as an oxidant. The nitric acid is regenerated within the reaction vessel in a manner similar to the high temperature Redox process (Van Weert et al.). 3.1.5 Chlorine/Hypochlorite Pretreatment Sodium and calcium hypochlorite and gaseous chlorine additions to carbonaceous ore slurries (or their percolation through heaps) provide a more oxidizing environment than simple pre-aeration. The intent is to destabilize any existing gold organic complexes that would result in soluble loss 19 (Ahmadiantehrani et al., Scheiner et al. 1968) or to deactivate carbon species that would compete for the gold cyanide complex (Ahmadiantehrani & Hendrix). There is evidence that some of the auriferous iron sulfides are oxidized (Bayrakceken et al.) and this could enhance gold liberation which is in fact seen as some gold dissolution in the oxidizing chloride environment. The exact mechanism of gold complex adsorption onto organic and graphitic carbon species is still somewhat problematical so that the effect of chlorine species oxidation of carbonaceous gold ore is not entirely predictable. The oxidation of pyrite by hypochlorite would proceed electrochemically and the overall reaction could be described by: FeS 2 + 7NaOCl + 2NaOH = FeCl 2 + 2Na 2 S0 4 + 5NaCl + H 2 0 (34) 3.1.6 Bioleach Oxidation Bacterial oxidation of pyrite and arsenopyrite is the major cause of acid rock drainage since the bacterial species responsible occur naturally in waste rock dumps and sulfidic tailings. Several commercial installations are now operating where the appetite of Thiobacillus ferrooxidans for pyrite or arsenopyrite is harnessed to provide a controlled oxidation of the refractory matrix. This results in gold liberation and improved gold recoveries in cyanidation (Brierly, Bruynstein, van Asweegan et al., Hackl, Harris & Brierly, Nagpal et al.). An advantage of bioleaching is in the selective oxidation of high energy areas at the faults and defects of a pyritic sulfide matrix. For ores where microfine gold particles (500A to 20u) occupy 20 these faults and discontinuities in the pyrite mineral, it is not necessary to achieve complete oxidation to get high gold extraction. In the case for arsenopyrite where the gold is finely disseminated throughout the matrix, or pyrite/arsenopyrite co-occurrences where often the gold is enriched in the arsenic containing mineral, it is fortunate that the bio-oxidation rate is enhanced. Often, nearly complete oxidation is possible in treatment times as short as seven days (Spencer & Budden 1990, Komnitsas & Pooley 1990, Komnitsas & Pooley 1991, Karavaiko et al.). The bacteria oxidize ferrous iron to ferric and sulfide sulfur to sulfate as a metabolic process. It is unclear as to the fate of elemental sulfur which results from the chemical leaching of the original pyrite feed by the bacterially produced ferric iron (Chaudhury et al.) although bacterially assisted oxidation is plausible. 4FeS2 + 150 2 + 2H 2 0 = 2Fe 2(S0 4) 3 + 2H 2 S0 4 (35) Fe 2(S0 4) 3 + FeS 2 = FeS0 4 + 2S° (36) 2S° + 30 2 + 2H 2 0 = 2H 2 S0 4 (37) Other bacterial species such as Thiobacillus thiooxidans and higher temperature Sulfolobus acidocaldarius, have been tested for their amenability to refractory gold ore treatment (Liu et al.). The optimization of bioleach methods is empirical due to difficulties in direct bioassay determinations since the techniques for speciation and culture population determination are not as yet well defined. Visual enumeration, protein assay and infrared methods are in development (Ralph). In amenability testing, the variables of concern are culture type, nutrition requirements, 21 oxygen transport, temperature and time of leach to effect gold liberation. The measureable variables are mixing power input, pH, potential, pulp density and dissolved metal content. It has been suggested that careful control of pH is necessary to avoid precipitation of basic iron/arsenic/sulfate compounds (Barrett & Hughes) which with other jarosite formation can interfere with gold liberation. Bioleach treatment of carbonaceous refractory gold ore does not always address the preg-robbing problem. Newmont has developed and patented the use of species of the Pseudomonas, Achromobacter and Arthrobacter genera to modify or block the gold adsorption in active carbonaceous gold ore (Brierly & Kulpa). The use of diesel oil, ostensibly as a nutrient source for the bacteria in the Newmont process, has been used on its own as a carbon blinding agent. 3.1.6.1 Bioheapleach Cyanide heapleaching technology for low grade oxidized ores is well established (McLelland et al., Pantelis & Ritchie). The use of bacterial agents in a pre-oxidation step for low grade refractory ore is a natural extension of uncontrolled oxidation exhibited in sulfidic waste rock and tailings. The controlled process generally requires an acidification and nutrient addition to get the bacterial growth phase started and to enhance overall kinetics. As in tank bioleaching, the Thiobacillus ferrooxidans and other chemoautotrophs derive sustenance from the sulfidic matrix. The result is liberation of the occluded gold by oxidation and dissolution of pyrite (Chaudhury et al.). The use of cyanide as lixiviant is complicated by the iron compound precipitation upon neutralization to a pH amenable 22 to cyanidation. There has been some test work to examine the use of thiourea at low pH in this application. The use of bioheapleach with sulfide ingesting species will not completely relieve carbonaceous ores of a tendency to preg-rob (Afenya). The design of the heap to maximize solution percolation and encourage bacterial activity is the subject of a number of patents. Agglomeration using an inoculum provides an effective distribution of bacteria and improved percolation for cyanide, thiourea or thiosulfate extraction methods (Brierly & Hill). 3.2 Lixiviation The recovery of gold from refractory ore would ideally involve the use of a solvent that extracts only the gold, leaving the gangue intact. Indiscriminate solvents such as aqua regia lead to complicated purification requirements. The most common lixiviant for the dissolution of gold is cyanide, usually as a sodium or potassium salt. Other gold complexing agents have been developed to avoid the problem of cyanicides, to provide additional oxidizing potential to liberate and dissolve occluded gold, and to dissolve gold in the acid environment created by some of the stronger oxidizing pretreatments. Other developments in lixiviants are focused on combining the pre-oxidation and dissolution steps into one operation by using oxidizing species capable of gold complexation in solution. Cyanide gold extraction provides the benchmark for evaluating refractory ore treatments. 23 3.2.1 Cyanide Lixiviation The accepted mechanism for the dissolution of gold with cyanide is thought to proceed with the formation of a peroxide intermediate. 2Au + 4NaCN + 0 2 + 2H 2 0 = 2NaAu(CN) 2 + 2NaOH.+ H 2 0 2 (38) 2Au + 4NaCN + H 2 0 2 = 2NaAu(CN) 2 + 2NaOH (39) 4Au + 8NaCN + 0 2 + 2H 2 0 = 4NaAu(CN) 2 + 4NaOH (40) Once the appropriate pre-oxidation steps have been taken to liberate refractory gold it is possible to achieve a dissolution rate of 1.8 mg cm"2 h' 1. In the case of very fine gold liberated from a sulfide matrix the extraction rate is enhanced by the large surface area of the exposed gold particles (Haque 1992). The cyanidation stage is sometimes simply limited by kinetics where the gold surface is in some way passivated or where cyanide is consumed by base metal complexation. Methods of enhancing the dissolution kinetics are the use of intensive cyanidation techniques such as a pressurized leach or high concentration cyanide solutions (Kontopoulos & Stefanakis 1990). 3.2.2 Pressure Cyanidation A pipe reactor has been installed by Consolidated Murchison Mine to treat an antimonial gold concentrate. The increased oxygen activity at pressures as high as 13.5 MPa prevents the oxygen consumers present from interfering with cyanidation. Temperatures up to 60°C are shown to improve 24 the kinetics of cyanidation whereafter decomposition of cyanide becomes a concern (Davis & Patterson, Muir et al.). The method has limited use where the gold is occluded in a sulfide matrix since much of the original sulfide remains intact at the low temperatures and moderate pH of this operation. 3.2.3 Thiourea Lixiviation The use of thiourea for the direct extraction of gold has many intrinsic advantages over cyanidation of refractory ores where cyanide consumption is the major problem. A n important advantage comes from its enhanced kinetics, said to be 10-12 times greater than cyanide (Charley). The thiourea is oxidized by an added oxidant such as ferric iron or hydrogen peroxide to formamidine disulfide which serves as a selective oxidant for gold or silver. 2CS(NH 2) 2 + 2Fe 3 + = NH 2 (NH)CSSC(NH)NH 2 + 2H + + 2Fe 2 + (41) 2Au° + 2CS(NH 2) 2 + NH 2 (NH)CSSC(NH)NH 2 + 2H + - 2Au(CS(NH 2) 2) 2 + (42) Au° + Fe 3 + + 2CS(NH 2) 2 = Au(CS(NH 2) 2) 2 + + Fe 2 + (43) The major advantage of thiourea is that it dissolves gold in acidic solutions and may be used to treat the pre-oxidized acid pulps without a neutralization or solid-liquid separation stage (Deschenes, Gabra, Hiskey 1983, Raudsepp & Allgood, Yen & Wyslouzil). In general thiourea has found limited commercial application in part due to the difficulty in potential control to prevent the irreversible oxidation of the formamidine disulfide to elemental sulfur resulting in reagent consumption. There 25 is some indication that thiourea is carcinogenic and this has led to prohibition of its use in some countries (Yen & Pindred 1989). 3.2.4 Chlorine The use of chlorine for the dissolution of gold was known before the advent of cyanidation. Dissolved chlorine, hypochlorous acid and hypochlorite ion are all capable of oxidizing gold in a chloride medium. The predominant complex is AuCl 4 -(Figure 1.). The aurous complex, AuCl 2 _ can exist in high chloride activity solutions (Hiskey & Atluri 1988) but tends to disproportionate to gold metal and the auric complex. Pressure leaching with hypochlorite in a chloride solution followed by a hot hydrochloric acid leach extracted in the range of 75-98% of the gold from an ore for which cyanidation recovered only 20% (Yen & Pindred 1989), but hypochlorite consumption was prohibitively high for the best gold recovery. This is a general problem with chlorine systems since the potential is higher than necessary to effect gold dissolution and pyrite oxidation. The nature of the leach solution prevents the use of inexpensive conventional materials for leach vessels often requiring all wetted surfaces to be made of titanium. 26 3.2.5 Bromine The ability of bromine to leach gold was the subject of a patent in 1881 but like chlorine leaching, it was supplanted by cyanidation. The development of the K-process (assisted by the involvement of the Dead Sea Bromine Group) and the manufacture of bromo-hydantoins by Great Lakes Chemicals has lead to renewed interest in bromine gold leaching. The K-process has been said to leach gold at rates up to 100 times that of cyanide (Capps & Cameron) while avoiding the dissolution of copper. The process of gold dissolution is enhanced by the presence of both hypobromous acid and bromide according to: 2Au° + 3HOBr + 5Br + 3H + = 2AuBr4" + 3H 2 0 (44) The potential-pH diagram for the gold-bromide system predicts a stability for the AuBr 2" species as well (Figure 2.). Tests with bromo-hydantoins as a source of hypobromous acid for gold dissolution used a roasting oxidation pre-treatment to avoid exorbitant reagent consumptions (Dadgar). As in cyanidation, the presence of oxidizable substances such as humic acid or soluble sulfides leads to high reagent consumptions (Sergent). 3.2.6 Iodine The use of iodine as a gold lixiviant is the subject of a patent (McGrew & Murphy) and the applicability to refractory gold ores is the case where mild oxidation conditions are adequate to 27 liberate the gold. The I37I* couple at 103 M provides a potential of 0.71 volts which is sufficient to oxidize gold in the presence of iodine (Hiskey & Atluri 1988). The complex formed is the most stable of the halides examined and exists at lower potentials and higher pH's (Figure 3) than either the chloro- or bromo-gold complexes (Tran & Davis 1991). 3.2.7 Other Lixiviants There are several other oxidants or complexing agents for gold that operate in aqueous solution at a variety of potentials and pHs. The sulfoxy species such as the strongly oxidizing Caro's acid, (Lakshmanan), and the complexing lixiviant thiosulfate (Kerley 1981 & 1983, Langhans et al.) have been applied to various low grade or refractory gold ores. Sulfurous acid formed by the addition of sulfur dioxide is the subject of a patent claiming the formation of a soluble gold sulfite complex that can be adsorbed by resin in the presence of a chelating compound (EDTA) and air as an oxidant (Touro et al.). Malononitrile, a dinitrile of formula CH2(CN)2 and its many derivatives, can complex gold in a manner analogous to thiourea without the problem of reagent oxidation to elemental sulfur (Heinen et al. 1970, Scheiner & Lindstrom). Table 3 summarizes the optimal pH and potential regimes for the production of stable gold complexes for a variety of lixiviants. Figures 1-3 demonstrate the gold halide stability areas in the form of Eh-pH diagrams (Hiskey & Atluri). 28 Table 3. Gold Lixiviants Lixiviants Complex PH E h V S H E Cyanide Au(CN) 2" 10-11 -0.6 Thiourea Au(CS(NH 2) 2) 2 + 2 0.4 Chloride AuCl 4 " <3 1.0-1.2 Bromide AuBr 4" 4-6 0.7-0.9 Iodide Aul 4 " 4-10 0.7 Thiosulfate Au(S 2 0 3 ) 2 3 - >7 0.1 Polysulfide AuS- > 10 0.5 29 30 31 3.3 Adsorption The use of carbon-in-leach (CIL) and carbon-in pulp (CIP) is well established (Ibrado & Fuerstenau). These techniques provide a large active surface area for the adsorption of the aurocyanide complex and can compete with carbonaceous refractory ore preg-robbing constituents (Guay 1981, Nendick). CIL has been effective where the refractory ore contains active carbon by competing for the gold complex as it is formed. Both techniques are useful when humic or fulvic acids are the problem since these vegetable source acids may complex gold and are in turn adsorbed by the activated carbon (Lee etal.). Alternate competing adsorbents for the recovery of gold from carbonaceous gold ore would include various ion exchange resins suitable for use in resin-in-leach circuits. The resins are not prevented from use in alternate lixiviant systems since thiourea and halogen complexes can be adsorbed as well. 3.3.1 Resin-in-leach Strong base quaternary amines have been tested for use in treating double refractory ores where cyanide is used as a complexing agent (Hiskey & Atluri 1989). The results suggest that the resin competes effectively with the carbonaceous preg-robbing component of the ore and only the portion of the gold that remains uncyanidable is lost to tailings. Several resins are available for gold adsorption and can likely accomodate the various natures of the oxidized gold complex solutions that 33 may result from ore specific treatments (Green et al.). This would indicate the use of competing resins in slurries where the oxidative method used to liberate gold from a sulfide matrix has activated a carbonaceous component causing preg-robbing. 3.3.2 Carbon-in-Chlorine Leach Research done by the U.S. Bureau of Mines examined the carbon-in-chlorine leach (CICL) concept to combine an oxidative pre-treatment with lixiviation and recovery from solution (Greaves et al.). The conclusion was that some sulfide oxidation and preg-robbing deactivation can take place in the chlorine leach but that the carbon adsorption is not efficient in the uncontrolled potential regimes experienced. It may be possible to design a process where the potential and pH is optimized at the end of the leach stages to maintain the chloro-gold complex in solution until contacted with carbon in a carbon-in-chlorinated pulp (CICP).This would conserve oxidizing potential for the leaching stage and avoid the consumption of oxidant in the adsorption stage. 34 4.0 S C O P E A N D O B J E C T I V E S The description of a gold ore as refractory depends simply on the ease by which its precious metal content can be extracted with conventional cyanidation. The common benchmark used by industry is 80% gold extraction. This means that the reasons for which gold is not extractable can still be present and account for the remaining 20%. The literature search identifies many reasons for gold retention. It is reasonable for the ore processor to assess treatments on the basis of economic feasibility in the pursuit of higher gold extractions. The assessment process can involve extensive testwork of any of a variety of treatment processes discovered in the literature search. It is apparent that different processes can apply to different ores even those within the various zones of a single orebody. The following testwork will attempt to develop diagnostic approaches to differentiating the multiple reasons for refractoriness in a selection of sulfidic and doubly refractory carbonaceous ores. The approach uses a comparison of gold extraction by cyanide to determine the effectiveness of any particular treatment. The conditions under which the treatments are applied are derived from the literature and are not generally constrained by reagent consumptions nor are they optimized except in a general fashion. The treatments include pyrometallurgical and hydrometallurgical pretreatments for gold liberation as well as gold dissolution tests with various lixiviants. 35 5.0 R E S E A R C H R E S U L T S A N D DISCUSSION 5.1 Feed Characterization The ore received for the purposes of experimentation was represented by drillcore, blasthole and millfeed samples exhibiting various degrees of refractory behaviour. The samples represented two distinct minesites and encompassed varied subtypes from those locations. The characterization of the samples involved analysis of the pertinent elements, some chemical compound approximations by assay and inference, a consideration of mineralogy, an examination of physical state and finally diagnostic leaching to estimate the level of refractory behaviour and reagent consumption. From this feed characterization an experimental plan was developed to examine the response of the varied ore types to a variety of conventional and novel treatments intended to improve the extraction of gold. 5.1.1 Sample Identification The first sample received came from the Cordex Exploration Company and was identified as drillcore D-231 from the Secret Pass zone of the Daisy property in Nevada. It was divided into selections of five foot mineralized intersections of drillcore from the 340 to 410 foot range. Further samples came from the Pinson Mining Company in Nevada and included mill feed from operations on June 9,1993 and two blasthole samples identified as B H 2841 and B H 2844. 36 The samples were sorted and five were chosen for treatment on the basis of inferred mineralogy and grade of precious metal with overall consideration for the refractory nature of the ore. The Secret Pass drillcore represented a very refractory, low carbonate, low organic carbon, moderately high sulfide content ore. The two mineralized intersections chosen for treatment represent low and higher grade gold samples of the overall deposit and are identified in Table 4 by their drillcore interval (SP340, SP405). The Pinson mill feed is a composite sample containing high carbonate carbon, significant organic carbon, somewhat less sulfide than the drillcore samples and exhibits mildly refractory behaviour (Pinson). The blasthole samples have varied carbon levels with significant carbonate, moderate sulfide, widely differing gold contents and moderate refractory behaviour (2841, 2844). The carbon assays were performed with commercial Leco induction furnace equipment using infrared detection for total carbon and carbonate carbon with the difference assigned to organic carbon. Total sulfur and sulfate sulfur were determined gravimetrically. Sulfide sulfur was determined with the use of a Leco induction furnace with a titrimetric analyzer using weak acid leached (sulfate depleted) samples. 37 Table 4. Refractory Ore Head Analysis Sample Carbonate Organic s2- S 0 4 S total C N CIL Carbon Carbon Extract Extract % % % % % % % SP340 - 0.01 3.83 0.18 3.89 2.6 SP405 - 0.07 3.69 0.15 3.87 13.3 Pinson 2.92 1.35 1.33 - 1.35 71.8 2841 1.52 ,3.78 1.45 0.03 1.43 74.7 79.2 2844 1.53 0.64 1.20 0.09 1.28 66.5 67.0 The cyanidation results (CN Extract) and cyanidation with carbon results (CIL Extract) are from diagnostic cyanide leaches with five gram per litre sodium cyanide solutions. The limited extractions confirm the refractory nature of the ores although only two can be considered extremely refractory. 5.1.2 Head Sample Analysis Sample analysis involved blending and riffle splitting of the finely crushed bulk samples into one kilogram charges for subsequent rod mill grinding. Further splitting of these minus 3/8 mesh feed samples produced 250 gram splits which were pulverized with steel media and analyzed by conventional means with highlights of the results recorded in Table 5. 38 Table 5. Elemental Ore Analysis Sample Ag As Au Ca Cu Fe Hg Mg Sb V ppm ppm ppm % ppm % ppm % ppm ppm SP340 0.4 728 1.3 0.28 29 4.19 1 0.04 6 4 SP405 1.6 1195 3.6 0.48 30 3.9 3 0.09 12 10 Pinson 0.8 208 4.49 2.28 94 2.06 11 0.06 6 59 2841 0.6 148 2.13 6.07 573 1.75 5 0.09 4 38 2844 0.6 602 6.99 2.41 123 2.15 17 0.64 8 41 From the elemental analysis of the feed samples by inductively coupled plasma (ICP), some conclusions could be drawn as to the nature of the ore mineralization. In all cases, silver represents less than twenty five cents per tonne of ore and is of minor economic consideration. The arsenic may provide a clue as to whether the presence of arsenopyrite contributes to the low gold recoveries. Assuming that all the arsenic exists as arsenopyrite leads to the results recorded in Table 6 which indicate that only a small percentage of the sulfide sulfur can exist as arsenopyrite. It is plausible that the gold can be associated with an arsenious sulfide phase if the relationship seen in some South African ores is duplicated (Claassen). The gold values in the various ores range from $15-88 US per tonne (at $300 US per oz.) and this suggests that a low cost, simple process will be required to economically extract the precious metal values. The calcium content is an indicator of the potential for acid consumption in any low pH treatment when it is present as a carbonate and becomes a nuisance when it leads to the precipitation of gypsum in processes in which oxidation of sulfide to sulfate is necessary. Table 6 estimates the maximum calcium carbonate content of each ore. The copper content of these ores is generally low and should not contribute significantly to cyanide 39 complexation or oxidation and is likely to be of limited use in copper catalyzed processes. The mercury does not impede the recovery of gold to any great extent but becomes an environmental consideration in choosing an extractive process since levels in excess of 10 ppm in two of the samples will require attention in a commercial process. The magnesium content is often an indicator of dolomite and impacts a process where the rate of neutralization of acid is significant since dolomite is typically less reactive than calcite. The dissolved magnesium in hydrometallurgical processing of sulfides often contributes to the recirculating load of metals in any closed circuit acid process due to the solubility of the sulfate. The antimony levels are enriched with respect to the average earth's crust level of 1 ppm but are still low and would require an intensive search by microprobe to show any passivation of gold particle surfaces or any compound formation that may affect gold extraction with cyanide. The consequence of this evaluation is that the ores contain a number of elements that can contribute to refractory behaviour but they are present in relatively minor amounts. Table 6. Derived Mineralogy Sample FeAsS ^ FeAsS / S 2 C a C 0 3 max % % max % SP340 0.16 0.8 0.7 SP405 0.26 1.4 1.2 Pinson 0.05 0.7 5.7 2841 0.03 0.4 12.8 2844 0.13 2.2 6.0 40 5.1.3 Mineralogy The higher levels of contained sulfides in the drillcore samples permits an examination of mineralogy through the use of a Siemen's Kristalloflex X-ray diffractometer (XRD) which results in the scans appended as X R D SCANS. The method suffers from a lack of resolution for minerals present in amounts less than 5%. The major minerals proved to be quartz and potassium feldspar. The predominant sulfide is pyrite and due to the resolution limits no other sulfides are readily identifiable. A comparison of the SP340 and SP405 patterns shows the possibility of a clay component in the SP405, most likely kaolinite and clinochlore. A sample of the SP405 ore was ground to a P g 0 of 78 microns in a steel rod mill and a stirred slurry cut was taken for examination under a Carl Zeiss reflecting light microscope. The sample preparation involved panning out the higher density minerals to produce a gravity concentrate of about 5 grams from a 50 gram sample. This concentrate was then mounted in epoxy and polished in stages ending in a one micron diamond paste polished finish. Examination at 128x magnification confirmed the presence of pyrite, quartz and silicate minerals such as potassium feldspar. There was minor evidence of hematite rimmed pyrite indicating that there had been only limited oxidation and this may indeed have been an artifact of the sample preparation. Infrequent occurrences of well polished whitish grains that could be arsenopyrite were encountered. The pyrite existed as mostly liberated anhedral and euhedral grains but there was some evidence of smaller pyrite grains trapped in silicate gangue. Any gold associated with the encapsulated pyrite would be unavailable for cyanide extraction. 41 5.2 Pretreatment 5.2.1 Experimental Design To test the amenability of oxidative treatments for the modification of refractory gold ores it is reasonable to first apply conditions that are, as much as possible, unconstrained by reagent consumption or kinetics to help identify the types of refractory behaviour. Physical occlusion is reduced by grinding to a practical degree of fineness generally 80% passing a 200 mesh (74//) screen. A highly oxidizing pressure leach of the Redox variety ensures sulfide oxidation is complete and provides a comparison for the likely more feasible partial oxidation treatments. Carbon activity and the degree of preg-robbing responds variably to hydrometallurgical treatments but can be eliminated in a dry thermal pretreatment or oxidizing roast. A variety of possible approaches for the treatment of the ores was chosen for testing and is represented in Figure 4. Standard ore preparation methods provided feed stock for the various pretreatments. Comparison of the effectiveness of the pretreatments was made using a diagnostic cyanidation method that somewhat masked the effect of residual cyanicides by using a generous dose of cyanide. This largely determined the comparative degree of gold liberation for the pretreatment methods. Alternate lixiviants including thiosulfate and the halides were tested for gold extraction from ground ores, oxidized residues, and in some cases the oxidation pretreatment pulps. This was 42 to aid in the investigation of combined oxidation and lixiviation in a continuous process without filtration. Stirred tank tests were conducted using a variety of oxidizing agents to reduce the activity of carbonaceous fractions and liberate gold from the sulfide matrix while removing oxygen consumers. Some methods provide oxidation potential without direct complexing or lixiviating ability (Redox, peroxide, persulfate) while others combine the oxidation reaction for sulfides with the ability to directly complex in-situ liberated gold (hypochlorite/chloride). The general procedure for the hydrometallurgical leach tests was to agitate slurries adequately to ensure solids suspension. This was typically 250-400 rpm for both Parr autoclave tests and glass resin kettle or beaker tests where solids were suspended but splashing was avoided. Acid additions generally started with any indicated sulfuric acid first followed by heating to the desired temperature and then adding the indicated oxidant. Nitric acid addition to high temperature tests was done through a pressurized stainless steel 'bomb' through which oxygen could be passed to ensure full reagent addition. Nitrogen and oxygen used for headspace makeup or air displacement were medical grade. Acid consumption, when quoted on a weight to weight basis with respect to ore, refers to bottle strength acid. 43 Ore Characterization and Preparation Pyrometallurgical Treatment Hydrometallurgical Preleach Bioleach Persulf ate/Peroxide Hypochlorite Leach Sulfur Dioxide Leach Nitric Acid Leach Gold Lixiviation Bromine Cyanide Bromine/Chlorine Thiosulfate Carbon-in-Leach Figure 4. Experimental Plan 44 5.2.2 Grinding The ore samples were cone crushed to one quarter inch size and then split into 1 kilogram charges for wet grinding. A steel mill with an approximately 30 kilogram rod charge was used to grind one kilogram charges of each of the ore samples at 66% solids. The ground pulps were washed out of the rod mill using Vancouver tapwater, settled and decanted to approximately 50% solids. Duplicate stirred samples were analyzed for particle size and the remaining slurry stored in sealed containers. The target was to achieve the common mill grind of 80% passing a 200 mesh Tyler screen. The results in Table 7 are averages of the indicated test grinds and on the whole ranged from 69 to 84% passing a 200 mesh screen for the samples used in the testwork. The one exception is the intentionally fine grind of the SP405 ore (G9) used to compare mechanical liberation with the oxidation methods. Table 7. Grind Data Test Feed Grind Time 1*80 -#200 Au # min % g/t G l SP340 15 131 63.0 1.3 G2,G3,G4,G18 SP340 20 88 69.0 1.3 G5,G7,G17 SP405 20 78 76.9 3.6 G9 SP405 30 37 96.2 3.6 G8,G10 Pinson 20 71 81.7 4.5 G11,G13,G15 2841 20 63 85.0 2.1 G12 2844 20 50 90.0 7.0 G14,G16,G19 2844 18 62 84.0 7.0 45 The fine grinding of ores does not always result in an improved gold extraction. The sub-micron gold in many refractory ores remains occluded by the sulfide matrix or gangue minerals. There is the further complication of adsorption of dissolved gold on gangue minerals, particularly clays that can provide adsorptive surfaces for the gold complex. A l l of the direct cyanidations are summarized in Table 8. The tests were performed at 6% pulp solids and 5 gram per litre sodium cyanide additions. Table 8. Cyanidation of Ground Ore Feed Feed P80 Test C N Ext type grind # % SP340 G4 88 CG4 2.6 SP405 G5 78 CG5 13.3 SP405 G9 37 CG9 22.9 Pinson G8 71 CG8 71.8 2841 G i l 63 CG11 74.7 2844 G12 62 CG12 66.5 Grinding the SP405 ore to a P 8 0 of 37 microns gained a further 9.6% gold extraction. This is not likely to improve to a feasible level of extraction by further grinding alone. 5.2.3 Diagnostic Lixiviation Direct cyanidation of the various ores using a diagnostic procedure confirmed their refractory nature. The cyanidation conditions used throughout this testwork were 5 grams per litre of sodium cyanide at 6% solids density and pH maintained between 10.5 and 11 with 10% lime slurry with any variations noted along with the extraction results. The length of the tests ranged from 18 to 22 hours 46 depending more on the time to achieve a steady pH in excess of 10.5 than concern about the rate of gold extraction in the last few hours of the leach. Carbon-in-leach (CIL) tests used 1.0 dry gram (typically 33 kg/tonne of ore) of Norit extruded peat source carbon rods added in a water saturated form to ensure activity. These rods resisted attrition and were easily removed from the leached pulps with a 20 mesh screen. These leaches could also be referred to as carbon-in-pulp (CIP) tests when carbon additions occur sometime after the cyanide is added. The active carbon addition process was used to compete with natural preg-robbing components, particularly in the organic carbon containing ores, and provides a qualitative assessment of adsorptive activity. The summary of unoxidized ore tests in Table 9 suggests that marginal improvement in gold recoveries can be expected for the 2841 and 2844 ores with the use of active carbon in the leach. Table 9. Ore Cyanidation Reagent Consumption Feed Test C N Cons. Lime Addn. C N Ext. CIL Ext. type # kg/tonne kg/tonne % % SP340 CG4 31.4 4.4 2.6 SP405 CG5 31.9 1.5 13.3 Pinson CG8 24.3 2.2 71.8 2841 CG11 14.0 2.2 74.7 2841 CIPG11 16.6 2.2 79.2 2844 CG12 12.5 2.2 66.5 2844 CIPG12 11.9 2.9 67.0 The high cyanide consumptions (CN Cons) for all of the tests on untreated ore are partially an artifact of the test conditions since the solution cyanide strength is high for the pulp density used. 47 The consumptions are determined by titrating the free cyanide and weak acid dissociable cyanide availale at the end of the test with silver nitrate. These conditions were used to assure residual free cyanide for overnight tests. The exceedingly high cyanide consumptions for the SP340 and SP405 ores reflect their high sulfide content and indicate that cyanide is being complexed by gangue mineral solutes, adsorbed by clay components and /or reacting to form cyanates and thiocyanates. The lime additions reflect the amount of 20% finely ground lime slurry that was used to ensure a pH in excess of 10.5 before cyanide was added to the leach. The stated quantities are reasonable and are useful as a baseline for comparison to the added expense engendered by acid pretreatment neutralizations. 5.2.4 Bioleach Oxidation The oxidation of the refractory gold ores by a mixed culture of Thiobacillus ferrooxidans, Thiobacillus thiooxidans and Leptospirillum ferrooxidans in both shakeflask and stirred reactor configurations was undertaken. Bioleach oxidation has been applied successfully in stirred tank reactors and is being developed as a low cost heap leach option. Shakeflask tests used 30 gram charges of the wet ground refractory gold ore feed, 100 grams of 9K nutrient solution as per the formula published by Silverman and Lundgren [1959] (Torma) sufficient quantities of 6M H 2 S 0 4 to lower the slurry pH to less than 2.7 and a measured charge of inoculum resulting in approximately 20% solids slurries. The primary inoculum consisted of mixed culture grown on gold-free pyrite concentrate in 9K nutrient solution at 10% solids. Sequential tests allowed 48 the use of active culture grown on the refractory gold ores themselves to inoculate subsequent tests. The tests were stirred and incubated at 35 °C in a rotary shaker. The oxidation-reduction potential (Eh) and the hydrogen ion activities (pH) were monitored using silver chloride reference electrodes with platinum or glass working electrodes respectively. Table 10 summarizes the measured variables of the shakeflask tests. Table 10. Shakeflask Bioleach Parameters Test Feed Feed Feed 9K H 2 S 0 4 Inoculant Time Temp pH pH Final Eh # type grind g 9 6M ml days C init. final mv SHE B1 SP340 G4 30 100 0.0 Pyr 12 35 2.13 1.52 733 B2A SP405 G5 30 100 0.0 Pyr+10ml B1 22 35 2.63 1.16 670 B3A SP405 G5 30 100 0.0 Pyr+10ml B1 22 35 2.63 1.19 673 B4 SP340 G4 30 100 1.0 Pyr 21 35 1.54 1.40 437 B5 SP405 G7 30 100 0.0 Pyr 26 35 1.78 1.03 738 B6 Pinson G10 30 100 0.3 Pyr 26 35 1.95 1.45 733 B7 2844 G14 30 100 2.0 Pyr 18 35 1.93 1.33 740 B9 SP405 G17 30 100 1.0 Pyr 54 35 1.37 1.27 695 B10 SP405 G17 30 100 0.0 Pyr 17 35 2.35 1.19 701 The first three tests treated the sulfidic ores and when the biological activity lag period reached 12 days for the higher grade (SP405) feed the tests were re-inoculated with the active culture growing on the lower grade sulfidic ore (test B l using SP340). Subsequent tests with the sulfidic ores were acidified to a pH below 2.5 before inoculation. Larger tests using 200 grams of refractory gold ore feed were performed in a covered, heated, one litre resin kettle with air injected below the agitator. Each test was inoculated with the entire pulp of a shakeflask test conducted using the same refractory gold ore feed. The larger size of the stirred-tank tests allowed samples to be taken at intervals to attempt to correlate residence time and the 49 measured solution parameters to improved gold extraction. Table 11 summarizes the measured variables of the stirred-tank tests. Table 11. Stirred Tank Bioleach Parameters Test Feed Feed Feed 9K H 2 S 0 4 Inoc. Time Temp PH Eh # type grind 9 g 6M ml days °C mV SHE B8 2844 G16 200 600 21.0 B7 0 35 2.65 B8.1 7 35 2.30 417 B8.2 20 35 1.99 638 B8.3 33 35 1.96 718 B11 SP340 G18 200 600 0.0 B10 0 35 1.94 B11.1 12 35 1.56 678 B11.2 19 35 1.30 730 The lag period varied for the ore types and even varied over the drill core intervals in these unoptimized tests. A l l samples demonstrated bioleach activity after lag periods of 2 to 11 days. Figure 5 shows a typical pH and solution potential response to bacterial activity. The diagnostic cyanidation results for the bioleached, filtered and washed residues are summarized in Table 12. 50 0 5 10 15 20 25 30 35 Time (Days) Figure 5. Carbonaceous ore stirred tank bioleach Table 12. Bioleach Results Test Feed Time Temp pH Eh CN Ext CIL Ext Fe # type days C final mV SHE % % ppm B2A SP405 22 35 1.16 670 98.1 B5 SP405 26 35 1.03 738 99.0 B6 Pinson 26 35 1.45 733 83.2 B8.1 2844 7 35 2.30 417 85.6 6750 B8.2 2844 20 35 1.99 638 86.0 4130 B8.3 2844 33 35 1.96 718 85.0 B11.1 SP340 12 35 1.56 678 94.7 B11.2 SP340 19 35 1.30 730 94.7 The SP405 ore contains the highest levels of arsenic at 1195 ppm and has the greatest chance of recovery improvement by bioleaching due to the preferential attack on arsenopyrite and the common association of gold with the arsenic in refractory gold ores. The ore contains insignificant carbon 51 and with no tendency to preg-rob the cyanide gold extractions are spectacular. Similarly the SP340 ore has an improvement in extraction from 2.6 to 94.7% at 12 days in the bioleach reactor. Pinson ore showed a moderate improvement from 71.8 to 83.2% gold extraction suggesting that optimization of the bioleach conditions may be necessary or that the carbon component is active and unaffected by the bioleaching pretreatment. The related 2844 carbonate ore showed significant improvement (from 66.7 to 85.6% gold extraction) after 7 days of leaching and achieving a pH of 2.3 and a potential of 616 mV SHE. These conditions suggest only partial oxidation of the pyrite. Further oxidation coupled with oxidation product precipitation over a further 26 days resulted in no significant change in gold extraction. The solution iron had dropped from 6.7 to 4.1 grams per litre by the 20th day and the pH had levelled out at 2. 5.2.5 Thermal Pretreatments Considerable use has been made of pyrometallurgical techniques in the treatment of refractory gold ore. The complete oxidation of both the sulfide and carbon components often eliminates all of the refractoriness of an ore, although occlusion of gold in dense hematite calcines is possible in roasts performed at temperatures greater than 800 °C (Osmanson). This is an example of process induced refractory behavior and can usually be solved by an attritioning or grinding treatment after roasting. Since properly designed roasters are very effective at converting refractory gold ore into a feed amenable for cyanidation, the feasibility of this technology is determined by environmental and to 52 a lesser extent, energy costs. The scrubbing of roaster gasses for the removal of S 0 2 is a problem associated with thermal treatments that have an oxidizing component. As well, any thermal process must address the volatilization of toxic metals such as mercury, selenium and arsenic. The required development of higher efficiency scrubbers and improved electrostatic precipitators is part of the environmental cost of roaster operation. A phenomenon observed in the non-oxidative roasting of auriferous pyrite and arsenopyrite is the migration of gold. Gold particles in the sub-micron size range coalesce and become detectible by scanning electron microscopy in the roaster product (Swash & Ellis). This would suggest that a non-oxidative thermal process may be successful in liberating gold from a sulfide matrix by either conversion of that matrix or effective liberation of the gold. The volatilization of sulfur from pyrite would result in a pyrrhotite or troilite product containing coarser gold particles. This new sulfide matrix is likely to be less refractory without further treatment but can also be dissolved readily in a mild acid leach prior to an alkaline cyanide gold extraction. To test this concept a tube furnace was built to allow roasting to be conducted in the absence of oxygen. Iron filings were added to possibly adsorb or combine with the liberated sulfur in the production of pyrrhotite from pyrite. The roast was conducted in alumina sample boats.The choice of temperature and roast time at temperature is guided by research that suggests pyrite rapidly converts to pyrrhotite under inert atmosphere at 600-650°C (Grimsey & Aylmore). Further, direct conversion of arsenopyrite and pyrite to magnetite or hematite does not allow gold coalescence and tends to occlude the fine gold. The best conditions would convert the auriferous sulfides to porous 53 pyrrhotite or troilite with a magnetite rim and coalesced gold blebs to provide leachable gold and reduced pyrrhotite reaction with cyanide. If sulfur is available as S 0 2 , troilite can convert to pyrrhotite at 330°C on cooling. Between 560 and 590°C pyrite converts to troilite under inert or low sulfur atmospheres (Hausen). Carbon content in the ore requires oxidation before appreciable sulfur oxidation occurs (Chen & Reddy). Roasting in the presence of lime at a temperature of 500°C results in the retention of sulfur as calcium sulfides, sulfites and sulfates while there is evidence that the transition to pyrrhotite still occurs (Hagni et al.). Organic carbon components including humates are mostly pyrolized by 350°C generating varying amounts of carbon dioxide and low molecular weight hydrocarbons depending on the available oxygen. The carbon dioxide of calcium carbonate begins to dissociate at temperatures as low as 550°C. This suggests that even dead roasts under inert atmospheres can remove problematic carbon content from the ore. Highly agitated oxidizing roasts are effective refractory gold ore pretreatments since the particle temperature is more closely controlled and kinetics are enhanced by better bulk gas/solid contact. 54 Table 13. Roasting Conditions and Results Test Feed Feed Temp Time Atmos Feed Iron Res Wt Loss CN Ext # source °C hr 9 g g % % T1 SP405 G7 500 4 N 2 30.00 3.13 32.62 1.54 52.4 T2 SP405 G7 640 4 N 2 30.05 3.00 32.37 2.06 50.5 T3 SP405 G7 600 19 air 41.00 0.00 39.58 3.46 73.3 T4 2844 G12 700 4 N 2 32.08 0.00 30.31 5.52 82.9 T5 2844 G12 720 18 N 2 32.55 0.00 30.76 5.50 81.5 T6 2844 G12 500 21 N 2 32.24 0.00 31.63 1.89 77.1 T7 2841 G11 750 4 N ? 32.47 0.00 29.39 9.49 66.2 Tests of low carbon /high sulfide ore (SP405) were examined for enhanced pyrrhotite production by attempting to capture the pyritic sulfur on finely divided iron at temperatures where pyrite converts to pyrrhotite. The samples were purged, roasted and cooled under a nitrogen atmosphere. Gold extraction from the 500°C test residue slightly exceeded that of the 640°C residue. The X-ray crystallographic analysis using a Siemens D5000 diffractometer did not find much alteration in the low temperature residue, but indicated significant pyrite loss and the presence of troilite in the 640 °C roast. Roasting SP405 for 19 hours at 600 °C in the presence of air resulted in much improved gold extraction and X-ray analysis indicated residual pyrite, a possible ferric arsenate and no troilite. Roasting the higher carbon/moderate sulfide ore (2844) for 4 and 18 hours under nitrogen at 700-720°C resulted in similar gold extractions but the 21 hour roast at 500°C suggests that carbon content impedes the beneficial effects of roasting at lower temperatures. A high temperature non-oxidizing roast of the high organic carbon ore (2841) actually reduces the gold extraction possibly due to hematite occlusion resulting from pyrite oxidation by small amounts of CO as well as the activation of remaining carbon components as adsorption sites. 55 Further investigation may reveal that a thermal test under a non-oxidizing atmosphere can distinguish between organic carbon refractory ores that are thermally activated, arsenopyrite/pyrite refractory ores and naturally preg-robbing ores. This can be useful where crystallographic analysis is inconclusive due to low mineral content. . 5.2.6 Nitric Acid Leach The ambient pressure stirred tank version of the Redox process was designed for application to low grade ores primarily as a sulfide oxidizing method. The typical procedure is to use stoichiometric quantities of nitric acid oxidant based on the sulfide content under an inert atmosphere to achieve nearly complete oxidation without redissolving N 0 2 formed in the headspace by contact of NO with 0 2 (Equation (27)). The intent is to achieve nitrate free pulps. The pressure leach of low grade ore in a titanium Parr autoclave at temperatures exceeding 200 °C is likely to achieve complete sulfide oxidation due to the recycling of the oxidation catalyst. The Redox process has variable results with active carbon. It is likely that humic substances are eliminated in the high potential, high temperature leach, but it is unlikely that oxidation of graphitic carbon is complete. Further, precipitation reactions can occlude precious metal as in the case of silver jarosite formation. High temperature autoclave tests are summarized in Table 14. The acids used were ACS reagent grade and are quoted at bottle strength. A two litre Parr autoclave with titanium wetted parts contained the reacting mixture. Oxygen overpressure of 100 psi above the steam pressure ensured 56 high oxidizing potentials. Samples were taken after rapid cooling and filtering of the residues and leached using the diagnostic cyanidation method. Table 14. High Temperature Redox Results Test Feed Solids H N 0 3 H 2 S 0 4 Time Temp PH CN Ext # % kq/t kq/t hr °C % R1 SP340 20 460 80 0.5 195 <1 94.7 R2 SP405 20 460 100 0.5 195 <1 99.0 The two autoclave tests at high temperature and extreme acidity were used to treat the very refractory sulfidic ores. The residues were readily cyanided to very low tails suggesting that silicate occlusion of gold is minimal and that the refractory behaviour is due entirely to sulfide occlusion since little or no carbon is associated with the ore. Table 15. Low Temperature Redox Results - Sulfidic Ore Test Feed Feed H N O 3 H 2 S 0 4 Time Temp PH Eh Solids CN Ext # Grind kg/t kg/t hr °C Pt/AgCI % % R3 SP340 G4 540 38 2.0 85 0.3 >800 26.4 97.4 R5 SP340 G4 540 38 2.0 85 0.2 900 39.2 97.4 R11 SP340 G18 540 38 0.5 85 0.0 715-800 28.5 97.4 R12 SP340 G18 540 38 1.0 85 -1.4 700-830 28.6 94.7 Low temperature, ambient pressure stirred tank tests using excess nitric acid to effect sulfide oxidation were universally effective at liberating the gold content. Although this process can generate elemental sulfur from arsenopyrite oxidation, the low arsenic content of the ore (728 ppm) ensures that the amounts would be small and according to these results, ineffective at occluding gold. 57 Sulfuric acid additions were based on acid consumption titrations to pH 1.5 at 60°C. This acidification ensured free acid availability in the early part of the leach. The low temperature Redox process leaches are conducted under a nitrogen blanket ostensibly to prevent re-entrainment of N 0 2 as nitrate but all of the low temperature tests used excess nitric acid based on sulfide stoichiometry. Although reagent intensive, the Redox process results define the maximum gold extractions that can be achieved with sulfidic refractory gold ores. Table 16. Low Temperature Redox Results - Carbonaceous Ore Test Feed TOC H N 0 3 H 2 S 0 4 Time Temp PH Eh CN Ext CIL Ext # % kq/t kq/t hr C Pt/AqCI % % R4 Pinson 1.35 188 74 2.0 85 0.6 790 84.7 90.1 R7 Pinson 1.35 188 73 1.5 85 <1 400-690 90.1 R7 Pinson 1.35 188 73 3.0 85 <1 400-690 84.0 88.5 R6 2841 3.78 225 149 2.0 85 1.0 735-775 70.2 72.6 R8A 2844 0.64 186 100 1.0 85 0.7 706 72.4 R9 2844 0.64 186 100 0.5 85 0.5 719 79.4 In leaching carbonate carbon containing ores in high acid, care must be taken to ensure adequate head space is available to contain the frothy pulp. The three carbon containing ores have similar sulfide content and are distinguished only by their range of organic/graphitic carbon (TOC) content. The Pinson ore is roughly the average of the other two and demonstrates refractory behavior midway between the others before treatment. The anomaly is that the lowest TOC containing ore (2844) is the most refractory. Behaviour such as this prevents simple carbon analysis from predicting the degree of gold retention. In general, shorter leach times in nitric acid pulps improved gold recovery 58 reflecting the destruction of the sulfide matrix effect on liberation. As leach times lengthen this ore type demonstrated reduced gold extractions by what appears to be a carbon activitation. Pinson Millfeed Redox 100.0 '. , , R , • 1 1 1 J 1 . 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 Pretreatment Time (hr) Figure 6. Carbonaceous ore showing evidence of process activation The nitric acid leach differentiates between sulfidic refractory and doubly refractory or active carbon containing ores. For sulfidic ores, gold that is not extracted after the Redox leach can probably be considered silica encased. For organic carbon containing ores, the lower extraction after longer leach times may be indicative of development of more active graphite or carbon rich organic components. Further examination of the carbon species after the Redox leach may clarify this relationship. 59 5.2.7 Sulfoxy Species from Sulfuric Acid, Peroxide and Persulfate The application of oxidizing potential to ore slurries by means of surrogates such as the nitrosyl ion in the Redox process can be extended to sulfoxy species. The thermodynamically predicted solution species resulting from sulfur oxidation are almost always sulfate in aqueous solutions due to the practically irreversable nature of sulfate production. In practice, a variety of metastable thio species exist in solution and can react with reduced sulfur present in the form of metal sulfides. The dissolution process is complex and is influenced by the conductivity of the sulfide, galvanic effects of mixed mineral phases and the genesis of metal depleted and passivated zones. The oxidation of pyrite is thought to proceed through a thiosulfate intermediate which can disproportionate to elemental sulfur and sulfate. The mechanism by which this is possible can be developed by considering the semiconducting nature of pyrite. Semiconductors are characterized by valence bands and conduction bands separated by a quantum mechanically forbidden energy band or, band gap. The energy required to promote an electron from the valence to the conduction band affects the mechanism of dissolution. Two types of charge carriers exist in semiconductors, electrons in the conduction band and holes in the valance band. The electronic structure of pyrite allows a crystal field interpretation of the iron to sulfur bonding scheme (Figure 7). It is proposed that pyrite resists dissolution because the upper energy levels of its valence band are occupied by non-bonding t2 g molecular orbitals. A band gap of 0.9 eV exists between this upper valance band and the anti-bonding eg* energy level of the conduction band. This 60 suggests that electron transfer to the conduction band, which is equivalent to a hole injection into the upper valance band, cannot contribute to bond breaking in the solid (Crundwell) since it is non-bonding orbitals that are affected. The consequence of the crystal field splitting is that pyrite is less easily oxidized. 61 4p 4s 3 d (jixrxiyyi)^ tcnxMxH) Fe ATOMIC ORBITALS u (TCP) /// F e S 6 MOLECULAR ORBITALS s i " ATOMIC ORBITALS (a) CONDUCTION BAND cr* (eg*) } A ~ l e V ' L2g VALENCE BAND (b) Figure 7. (a) Relationship between atomic orbitals, molecular orbitals, and the band molecular orbital diagram for the FeS6 cluster, (b) Band model for FeS2 (from Osseo-Asare) The dissolution of pyrite may begin with the injection of a hole into the t2g valance band by an oxidant. This hole may react with water at the surface: H,0 + h+ = FT + OH- (45) The hydroxyl radical formed is very reactive and the couple with the hydroxide anion has an estimated 2.95 volt SHE potential. This has sufficient energy to react with the bonding orbitals. 62 Mishra & Osseo-Asare propose the intermediate formation of an adsorbed thiosulfate species via a series of steps: FeS 2 ( S ) + H 2 0 + h + = Fe(OH)S 2 ( S ) + H + (46) This is a net adsorption of a hydroxide anion at a ferric iron surface site induced by the hole injection. This is followed by a migration to a sulfide lattice site: Fe(OH)S 2 ( S ) = FeS 2(OH) ( S ) (47) Surface hydroxylation: FeS 2(OH) ( S ) + 3H 2 0 + 3h+ = Fe(OH)2^S2(OH)2 + 3H + (48) is followed by oxidation of the sulfide: Fe(OH) 2«S 2(OH) 2 + 2h+ = Fe 2 + + S 20 3 2" •+ 2H + + H 2 0 (49) The net result is the production of an adsorbed thiosulfate which can disproportionate to elemental sulfur and a bisulfite anion: FeS 2 ( S ) + 3H 2 0 + 6h+ = Fe 2 + ( A Q ) + S 2 0 3 2 " ( A D S ) + 6 H + ( A Q ) (50) 63 + FT = S, '(ADS) + HSCy, 3 (AQ) (51) Since the thiosulfate can be oxidized to sulfate through polythionate and sulfite intermediates, this mechanism avoids the requirement of oxidizing any elemental sulfur to sulfate. If the thiosulfate formed by oxidation of pyrite by this route could be stabilized in solution then gold extraction could be effected by this in-situ lixiviant. The potentials measured for some sulfoxy and peroxide species suggest that there is sufficient oxidation potential to dissolve pyrite and change the nature of carbon species. The use of Caro's (peroxomonosulfuric) acid formed by the cold addition of hydrogen peroxide to concentrated sulfuric acid, has been examined as an oxidant for sulfides (Lakshmanan et al.) and the use of Oxone powder containing peroxomonosulfate has been studied as an oxidative enhancement to the cyanidation of auriferous pyrite (Stoychevski & Williams). The oxidizing strength of sulfur dioxide/oxygen mixtures in the presence of transition metal ions was the subject of a study done of high pH metal oxidations that postulated the formation of salts of Caro's acid (Devuyst et al.). The reduction potentials for peroxydisulfate and hydrogen peroxide are adequate for the purpose of sulfide oxidation as follows: S 20 8 2" + 2H + + 2e" = 2HSCV E° = 2.12 V (52) S 2 0 8 2 - + 2e- - 2S0 4 2- E° = 2.01 V (53) 64 H 2 0 2 +2FT + 2e" = 2H 2 0 E° = 1.78 V (54) Hot concentrated sulfuric acid is an oxidizing agent according to: 5H 2 S0 4 + 2e" = H + + S 0 2 + 4HS0 4 ' + 2H 2 0 (55) The oxidation potential is lost in dilute solution. The addition of hydrogen peroxide to a sulfuric acid slurry may result in metastable sulfoxy species forming at the surface of a metal sulfide and leading to dissolution without elemental sulfur remaining to passivate the mineral surface. To test this concept, ore samples were leached in titanium Parr autoclaves where an overpressure of oxygen or nitrogen in psi is indicated in Table 17. Tests in contact with air were done at ambient pressure in a stirred glass vessel. Hydrogen peroxide was added as a 30% solution in a quantity designed to maintain a potential in the range of 650 mV based on contact tests with silica sand in sulfuric acid slurry. The potentials achieved in the stirred pulps at pH 0-1 were certainly adequate to oxidize pyrite but the reaction kinetics are considerably lower than the nitric acid system. Tests with the sulfidic ores (SP340, SP405) showed only minor improvement in gold extraction after a leach time of 45 minutes. The tests with Pinson ore established that gold extraction improvement was a result of the added oxidant and acidification, since an inert atmosphere test at 50°C lead to similar extraction improvements as those of a test under 100 psi overpressure of oxygen conducted 65 over 2 hours. Similar tests with 2841 ore actually showed little or no effect under oxygen overpressure and slight improvement under nitrogen after 4 hours of pretreatment. Table 17. Peroxide Oxidation Test Feed H 2 0 2 H 2 S 0 4 Atm Time Temp pH Eh C N Ext CIL Ext # kq/t kq/t psi hr C mvPt/AgCI % % P2 Pinson 150 300 O 2#100 2.0 50 0.8 513 82.4 88.5 P3 Pinson 150 300 N2#100 2.0 50 0.3 503 84.0 86.3 P4 2844 200 300 air 1.5 27 0.5 620-660 82.1 P5 2844 200 300 O 2#100 2.0 50 0.8 502 87.1 92.3 P6 2841 200 300 O 2#100 4.0 50 1.0 493 72.6 79.0 P7 2841 200 300 N2#100 4.0 50 0.9 492 79.0 83.9 P10 SP405 200 300 air 0.75 22 0.6-1.2 480-620 28.6 33.3 P11 SP340 200 300 air 0.75 22 0.3-0.5 540-630 18.4 15.8 P12 2844 200 100 air 2.0 22 1.1-1.5 480-636 82.5 P15 2844 200 300 O,#100 2.0 80 <0.4 510 86.8 A series of tests with the most refractory carbonaceous sample (2844) provides a cursory examination of the effects of temperature on gold liberation. The dotted line in Figure 8 indicates 66 100 u 10 5 65 50 -' : 1 1 ! — i 1 20 30 40 50 60 70 SO Leach Temperature (C) Figure 8. C IL gold recoveries for 2844 ore after peroxide leaching the extraction without pretreatment. Peroxide leaching appears to be beneficial in short contact time tests (1.5 to 4 hours) with carbonaceous refractory gold ores. The acid addition was reduced to 100 from 300 kilograms per tonne in a 2 hour test in contact with air and the extraction by CIL was not changed significantly. 5.2.8 Oxidation with Hypochlorite in Brine Chlorination of gold ores was the primary recovery technique before the advent of cyanidation. Roasting ores with chloride salts lead to chloride volatilization methods, leaching in high potential 67 brines with chlorine formed soluble gold chloride complexes, pressure leaching with hypochlorite and oxygen as a pretreatment before cyanidation and mild chlorine/hypochlorite pre-aeration to destabilize organic gold complexes in carbonaceous ores have all been developed on an ore specific basis. The use of chlorine to remove the sulfur content of coal is an effective example of pyrite oxidation. A l l of these methods suffer from loss of oxidant since the chlorine is non-specific in the substances it will oxidize. Chlorine dissolves in water according to: In acid solutions the chlorine content is limited by the solubility of the gas, at high pH the chlorine is stabilized as the hypochlorite anion OC1 - . At high pH the metals are predominantly precipitated as their hydroxides or exist as oxyanions and are unavailable as surrogate oxidants. This suggests a process whereby the availability of chlorine oxidation potential can be combined with active metal oxidation reduction couples and likely increase the kinetics of oxidation of metal sulfides. A substance commonly used to retain 'free' chlorine in swimming pools at neutral pH is cyanuric acid. It exists as a triazine tautomeric ring structure with a stabilization analogous to a keto-enol formation. It has a greater resonance energy than a benzene ring and less tendency for the ring structure to enter into any reactions. Cyanuric acid or l,3,5-triazine-2,4,6(lH,3H,5H)-trione C l 2 + H 2 0 = HOC1 + FT + CI" (56) At least one pathway for the chlorination of pyrite is: FeS 2 + 70C1-+ 20H- = FeCl 2 + 2S0 4 = + 5C1" + H 2 0 (57) 68 ((CNOH)3) is characterized by the chemistry of the three functional groups attached to the C3N3 ring and is of interest in its chlorinated form. Treating ore in a pH range of 5 to 8 with chloride, hypochlorite and cyanuric acid is the subject of a patent (Baughman et al.). Treatment of the carbonaceous refractory gold ores occurred in contact with air at ambient pressure or in a Parr autoclave under 50 psi of oxygen overpressure as noted in Table 19. The oxidant for this series of tests was calcium hypochlorite added as a solid so that pulp densities could be kept at 25% solids. A l l of the tests were filtered and washed, followed by cyanidation, to gauge the effect of the high oxidation potentials observed. The recorded extractions do not separately account for gold complexed in the chlorination stage as they were based on pretreatment feed assays. Table 19. Chlorination of Carbonaceous Ore With Cyanuric Acid Added Feed Oxidant (CNOH)3 NaCI Additives Additives Time Temp pH Eh CN Ext CIL Ext kg/t kg/t kg/t kg/t hr C mvPt/AgCI % % 2841 50 8.0 400 NaOH 12.0 20.0 45 7.4-8.9 300-750 77.6 79.1 2841 50 8.0 500 0 2 50psi 18.5 22 5.9 340 74.2 79.1 2844 50 8.0 400 NaOH 4.3 21.5 48 7.1-8.5 490-890 77.6 2844 50 8.0 500 0 2 50psi 20.5 30 77.6 75.6 Pinson 50 5.0 100 NaOH/HCI 3/7.4 4.0 22 8.0 1000 77.9 Pinson 50 8.0 400 NaOH 6.4 22.0 48 6.8-8.0 400-950 80.9 Pinson 50 8.3 500 HCI/0 2 6.2/1 OOpsi 20.0 50 6.1-8.0 290 77.9 Minor improvements are seen in the two most refractory ores (2844, Pinson). The extractions by cyanide and CIL are similar for the residues where both methods were used suggesting that some preg-robbing activity has been negated. 69 Tests using similar conditions and reagent additions for the sulfidic ores were unsuccessful in improving the gold extractions. The results were insensitive to chloride addition, time in leach, temperature and oxidant (Table 20). Table 20. Chlorination of Sulfidic Ore With Cyanuric Acid Added Feed Oxidant Oxidant (CNOH)3 NaCI Additives Additives Time Temp pH Eh CN Ext kq/t kq/t kq/t kq/t hr C mvPt/AqCI % SP405 Ca(OCI)2 116 4.17 0 NaOH 17.5 2 22 8.0 900 24.8 SP405 Ca(OCI)2 123 4.17 200 NaOH 11.2 2 40 8.0 880 23.8 SP405 Ca(OCI)2 116 4.17 0 NaOH 13.0 2 55 8.0 870 29.5 SP405 Ca(OCI)2 46 8.33 0 NaOH 31.5 4 22 8.0 350 58.1 SP405 NaOCI 23 8.33 0 NaOH 10.7 4 22 8.0 920 39.0 SP405 Ca(OCI)2 50 8.33 333 NaOH/HCI 6.8/13 4 22 8.0 990 27.6 SP405 Ca(OCI)2 50 8.00 400 NaOH/HCI 6.7/7.5 22 22 6.5-8.0 1000 26.7 SP405 Ca(OCI)2 50 8.33 500 HCI/02 9.1/100psi 20 50 1.7-8.0 503 39.0 The one test with an interesting result was the 20 hour leach under 100 psi of oxygen overpressure at 50°C. The final pulp was measured at 503 mV and a pH of 1.7 suggesting acid generation that would be expected to come from sulfide oxidation. The poor extraction result would suggest that considerable elemental sulfur was also formed and re-passivated the contained gold. The cyanidation test reagent consumption of 21.2 kilograms per tonne of ore for sodium cyanide is high but not unusual for these tests. 70 5.3 Lixiviation 5.3.1 Thiosulphate The use of thiosulfate in the recovery of gold has been researched by many (Berezowski & Sefton, Langhans et al., Zipperian et al.) and is the subject of several patents (Berezowski et al. 1978, Kerley Jr. 1981 & 1983, Perez et al. 1987). Each patent further refines the operating conditions for a successful refractory gold ore treatment with the tendency towards higher pH as time passes. As the pH increases, so too does the influence of the ammonia component as a gold lixiviant. A major benefit of the use of thiosulfate is the observation that the highly negatively charged complex does not adsorb readily on active carbon. The immediate benefit is the elimination of preg-robbing characteristics for carbonaceous ores. The conditions suggested for the thiosulfate leach are likely to have a destructive effect on unstable sulfides and this may be adequate to solve the sulfidic gold occlusion effect in some doubly refractory gold ores. The reactions of interest in this system are: The cupric tetrammine complex acts as an oxidant for gold and the Cu7Cu 2 + takes part in an equilibrium of complexed species. 4Au + 8S 2 0 3 2 - +0 2 + 4H + - 4Au(S 20 3) 2 3- + 2H 2 0 (58) Au + 5S 2 0 3 2 - + Cu(NH 3 ) 4 2 + = Au(S 20 3) 2 3" + 4NH 3 + Cu(S 20 3) 3 5" (59) 71 2Cu(NH 3 ) 4 2 + + 6S 2 0 3 2 - + H 2 0 = 2Cu(S 20 3) 3 5" + 8NH 3 + 0.5O2 + 2H (60) Sulfite added to the system generates thiosulfate from soluble sulfide and possibly metal sulfide surfaces with the generation of elemental sulfur. 3S0 3 2- + 2S2" + H 2 0 = 2S 20 3 2" + 60H" + S (61) Although written here as an oxidation that removes gold lixiviant strength, the reduction potential for tetrathionate to thiosulfate is 0.08 V . 2S 2 0 3 2 " +0 2 + 2H 2 0 = S 4 0 6 2 - + 40H- (62) Reduction of tetrathionate to thiosulfate and sulfate generates acid and controls the loss of thiosulfate to oxidation which is unfortunately catalyzed by copper. S0 3 2" + 20H" + S 40 6 2" = 2S 2 0 3 2 - + S0 4 2" +H 20 (63) Test conditions suggested by the patents and other research (Hemmati et al.) appear to be somewhat ore specific with regards to thiosulfate consumption, but in general the kinetics are faster than cyanide, the temperature should not exceed 65 °C and copper should be present in the few grams per litre range for pulp densities ranging from 10-50%. The ammonium thiosulfate concentrations range 72 from 10 to 400 grams per litre. Equilibrium conditions that favour the gold thiosulfate complex require a pH less than 8.8 for solutions that are 1M for thiosulfate and ammonia (Zipperian et al.). A scoping test was conducted with 60 grams per litre ammonium thiosulfate, 1 gram per litre copper sulfate, 15% solids for 8 hours at 50°C with sulfur dioxide at or near saturation and using bottle strength ammonia (28% N H 3 ) to maintain a pH in the vicinity of 9 for the highest organic carbon containing ore in this study (2841). The result was a significant reduction of gold extraction from 79.2% for CIL extraction to less than 35%. Subsequent tests used pre-oxidized ore feeds (low temperature Redox or bioleach) to compare the sulfur dioxide saturated/ammonia pH control system for thiosulfate leaching to the more conventional cyanidations. Table 21. Thiosulfate Lixiviation Feed Preoxid" NHJS CuS0 4 NH4OH Solids Control Time Final TS Ext CN Ext type Method kg/t kg/t kg/t % hr PH % % 2841 none 350 20 247 14.1 S0 2 8 7.8-9 32.3 79.2 2841 Redox 350 20 143 14.3 S0 2 8 7.6-10.4 79.0 72.6 2844 Bioleach 700 40 848 7.4 Na 2S 20 5 3 9.1 87.7 85.0 2844 Redox 350 20 903 12.9 S0 2 8 6.4-9.1 77.5 72.4 2844 Redox 350 20 475 13.6 S0 2 6 8.81 85.6 79.4 SP340 Redox 350 20 903 12.9 SO, 6 7-8.3 97.4 97.4 Some slight gains in extraction were seen for all of the oxidized residues when thiosulfate was used as a lixiviant with the exception of the sulfidic residue for which there was not much room for. improvement. It is likely that the difference in the extractions between the cyanide and thiosulfate tests can be ascribed to residual carbon activity. 73 5.3.2 Halides One of the disadvantages of acid oxidations as a pretreatment for refractory gold ores is the necessity of neutralization or a solid /liquid separation before an alkaline cyanidation for gold recovery. The halides form stable complexes with gold at low pH and may be suitable for treating oxidized pulps directly. Bromine as a gold lixiviant has been tested using liquid bromine as well as l,3-dibromo-5,5-dimethylhydantoin which generates hypobromous acid in solution in combination with bromide salts (Equation (44)). For iodine it is the triiodide which can act as oxidant: 2Au +13-+1- = 2Auiy (64) Mixed halide systems have the economic advantage of replacing the more expensive iodide or bromide salts with chloride. A commercial mixed halide treatment is known as the INTEC Copper Process (Everett). Tests are generally shorter for gold extraction with halides due to enhanced kinetics. Two kinetic series are plotted in Figure 9 along with their cyanidation extractions before pretreatment. 74 100 1 Time (hr) Figure 9. Gold extraction at 12 kg/t Br 2 and 6 kg/t KBr. The results of these tests suggest that a leach time of 2 hours or less would be sufficient. Table 22. Mixed Halide Gold Extraction Feed Test Br2 KBr H 2S0 4 Kl NaCI Time Temp Final Final 3hr Ext type # kq/t kq/t kq/t , kq/t kq/t hr C PH mV % SP405 BR14A 50 20 62.8 3 40 1.3 579 94.0 SP405 BR14B 50 20 61.4 200 3 40 1.2 894 94.8 SP405 BR14C 50 40 82.6 200 3 40 0.9 937 96.5 SP405 BR14D 50 20 62.0 40 3 40 1.1 857 99.1 SP405 BR14E 50 20 71.2 3 40 1.1 880 95.7 At least one study suggests that oxidizing potential is lost in a mixed system of iodide and bromine by the formation of Br 2P (Pesic & Sergent) but test results using a Redox process oxidized residue 75 demonstrated that this combination provided the best extraction (Table 22 test BR14D). The last test of the series incorporated active carbon on the recommendation of one study (Pesic et al.) and the results were not significantly different than the test without carbon (95.7 vs 94.0% gold extraction). Direct bromination of Redox leached pulps for SP340 ore gave virtually identical results to the kinetic leach test for this sulfidic ore. The oxidized pulp was diluted to 15% solids and cooled to 50°C after 1 hour of low temperature Redox leaching. Both one and two hour bromine leach samples resulted in gold extractions similar to those seen for a filtered and washed Redox residue. Table 23. Bromine/Bromide Gold Extraction Feed Test Br2 KBr H 2S0 4 Time Temp Gold Extraction % Type # kg/t kq/t kq/t hr C 1hr 2hr 3hr 4hr SP340 BR12 12 6 3.7 4 40 89.5 89.5 89.5 SP340 BR15 10 5 38 3.5 50 90.0 90.0 The immediate benefit of these results is that there is the possibility of eliminating one filtration step where acidic oxidation of sulfide is indicated as the refractory ore pretreatment of choice. 76 6.0 C O N C L U S I O N S 1. The chemistry of refractory gold ores is made complex by their heterogeneity which is promulgated in two major refractory characteristics; occlusion and adsorption. 2. Occlusion can be physical or kinetic, and these in turn can be natural or process induced while adsorption is generally reversible /kinetic and is also natural or process induced. 3. Natural occlusion is a function of the geochemical genesis of the ore whereas process induced occlusion can be caused by dense oxide phase development in roasting, surface passivating film formation from incomplete oxidation of sulfide or precipitant formation in leaching, and/or passivation by adsorbed non-auriferous complexes in lixiviation. 4. Natural adsorption or preg-robbing is related predominantly to the complexing of gold by the graphitic or organic components of the ore before the ore is treated whereas process induced adsorption occurs when ore components become active due to increases in surface area or components are oxidized to an adsorption active state with respect to the lixiviant in use, leading to retention of the normally soluble gold complex. 5. The degrees of refractoriness in gold ore can reasonably be distinguished as either economical or uneconomical to exploit. 77 6. The heretofore static analyses and characterization tests for ores can be augmented with kinetic evaluations using standard conditions that can often apportion refractory behaviour into its individual causes: a) Bioleaching can determine the quantity of gold occluded by arsenopyrite and leachable sulfides that may be available for extraction by partial oxidation. b) Nitric acid (Redox) oxidation at temperatures higher than 185 °C can determine the amount of gold that remains occluded by acid insoluble silicates in sulfidic ores. c) Nitric acid (Redox) oxidation at 85 °C liberates sulfide occluded gold and can activate the non-hydrometallurgically oxidizable carbon in doubly refractory ores. d) Inert atmosphere roasting below 650 °C liberates gold that is occluded by a sulfide matrix as well as that adsorbed by humic substances but can create active graphitic adsorption sites. e) Thiosulfate leaching can potentially determine the propensity of an ore to re-adsorb gold complexes with cyanide, thiourea or the halides due to the low uptake of thiosulfate gold complexes by active carbon. f) Bromine or mixed halide lixiviation provides a quicker test for gold solubility than cyanidation due to enhanced kinetics and can be used as a rapid test for degree of gold liberation for partial oxidation pretreatments. 7. The plant metallurgist has available many diagnostic procedures to distinguish the mechanisms for the resistance of an ore to leaching and can use these to circumvent costly -78 empirical studies combined with sophisticated analytical techniques by the use of simple amenability tests. 79 B I B L I O G R A P H Y Afenya, P. M . , Treatment of carbonaceous refractory gold ores., Minerals Engineering, v 4 n 7-11, 1991,p 1043-1055. Ahmadiantehrani, M . , Hendrix, J. L . , Ramadorai, G., Hypochlorate pretreatment in heap leaching of a low grade carbonaceous gold ore., Minerals & Metallurgical Proc, Feb. 1991, p 27-31. Ahmadiantehrani, M . , Hendrix, J. L . , Hypochlorite pretreatment of a low grade carbonaceous gold ore, part II: Effects of agglomeration, temperature and bed height., EPD Congress '92, J. P. Hager ed., The Minerals, Metals & Materials Society, p 63-79. Barrett, J., Hughes, M . N . , Mechanism of the bacterial oxidation of arsenopyrite-pyrite mixtures: The identification of plant control parameters., Minerals Engineering, v 6 n 8-10, Aug-Oct 1993,p 969-975. Bartlett, R. W., Aeration pretreatment of low grade refractory gold ores., Minerals & Metallurgical Processing, v 7 n 1, Feb 1990, p 22-29. Baughman, D. R., Huff, R. V . , Axen, S. G., Process for extracting metal values from ores., U.S. Patent # 5,169,503, December 8, 1992. Bayrakceken, S., Yasar, Y. , Colak, S., Kinetics of the chlorination of pyrite in aqueous suspension., Hydrometallurgy v 25 n 4, June 1990, p 27-36. Bayrakceken, S., Yasar, Y . , Colak, S., Leaching of FeS in aqueous chlorine solution., Hydrometallurgy v 25 n 4, June 1990, p 357-365. Beattie, M . J. V . , Raudsepp, R., Sarkar, K. M . , Childs, A . M . , The Arseno process for refractory gold ores., Impurity Control Disposal, Proc. CIM Annu. Hydrometal. Meet. 15th, Montreal, Quebec, 1985. Berezowsky, R. M . G. S., Sefton, V . B. , Recovery of gold and silver from oxidation leach residue by ammoniacal thiosulphate leaching., Preprint: 108th A I M E Meeting, New Orleans, February 1979, Pp 17. Berezowsky, R. M . G. S., Sefton, V . B., Gormely, L.S., Recovery of precious metals from metal sulfides., U.S. Patent 4,070,182, January 24,1978. Berezowsky, R. M . G. S., Weir, D. R., Factors affecting the selection of pressure oxidation for the pretreatment of refractory gold ores., E & M J International Gold Expo and Conference, Reno, Nevada, September 1989. 80 Berezowsky, R. M . G. S., Weir, D. R., Pressure oxidation for treating refractory uranium and gold ores., 22nd Annual Conference of Metallurgists of the C.I.M.M., Edmonton, Alberta, August 1983. Box, J. C , Yusuf, R., Simulation of heap and dump leaching processes., Extractive Metallurgy Symposium, AusIMM, Melbourne, 1984, p 117-124. Brierley, James A., Biotechnology for the extractive metals industries. JOM v 42 n 1 Jan 1990, p 28-30. Brierley, J. A. , Kulpa Jr., C. F., Biometallurgical treatment of precious metal ores having refractory carbon content., U.S. Patent #5,224,493, September 14, 1993. Brierley, J. A . , Hi l l , D. L. , Biooxidation process for recovery of gold from heaps of low-grade sulfidic and carbonaceous sulfidic ore materials., U.S. Patent #5,246,486, September 21, 1993. Bruynesteyn, A. , Mineral biotechnology., Journal of Biotechnology v 11 n 1 Jul 1989, p 1-10. Budden, J. R., Spencer, P. A. , Use of bacterial oxidation of refractory ores and concentrates to improve precious and base metal recovery., Fifth AusIMM Extractive Metallurgy Conference. Publ by Australasian Inst of Mining & Metallurgy, Parkville, Aust., Oct. 1991, p 153-159. Capps, P. G., Cameron, P. M . , New technology in gold processing., Amdel Ltd., Townsville, Queensland, Australia. Charley, W. R., Hydrometallurgical extraction of precious metals using thiourea., Practical Hydromet '83,7th Annual Symp. on Uranium and Precious Metals., SME-AIME, Lakewood, Colo., 1983. Chaudhury, G. R., Sukla, L . B., Das, R. P., Utilisation of low-grade pyrites through bacterial leaching., International Journal of Mineral Processing 26, (3-4), July 1989, p 275-284. Chen, B. , Reddy, R.G., Roasting characteristics of refractory gold ores., Advances in Gold and Silver Processing Proc Symp GOLDTech 4. Publ by Soc for Mining, Metallurgy & Exploration Inc, Littleton, CO, US A . p 201 -214. Chryssoulis, S. L. , Cabri, L. J., Salter, R. S., Direct determination of invisible gold in refractory sulphide ores., Proc. of the Int. Symp. on Gold Metallurgy, CIM, Winnipeg, 1987, p235-244. Claassen, R., Mineralogical controls on the bacterial oxidation of refractory Barberton gold ores., FEMS Microbiology Reviews 11, 1993, p 197-206. 81 Crundwell, F. K. , The influence of the electronic structure of solids on the anodic dissolution and leaching of semiconducting sulphide minerals., Hydrometallurgy, v 21 n 2,1988, p 155-190. Dadgar, A. , Refractory concentrate gold leaching. Cyanide vs. Bromine., JOM v 41 n 12 Dec 1989 p 37-41 Davis, D. R., Paterson, D. B. , Practical implementation of low alkalinity pressure cyanidation leaching techniques for the recovery of gold from refractory flotation concentrates. Gold 100, vol 2 Extractive Metallurgy of Gold, Publ by South African Inst of Mining & Metallurgy (South African Inst of Mining & Metallurgy Symposium Series S8), Johannesburg, S.A., p 591-604. Deschenes, G., Literature survey on the recovery of gold from thiourea solutions and the comparison with cyanide., C. I. M . Bulletin, vol. 79, November 1986, p76-83. Devuyst, E.A. , Conard, B. R., Ettel, V . A. , Pilot plant operation of the INCO S 0 2 / Air cyanide removal process., preprint 29th Ontario Industrial Waste Conference, June 13-16, 1982, pp21. Devuyst, E.A., Mosoiu, A. , Krause, E., Oxidizing properties and applications of the S0 2 -0 2 system., Hydrometallurgy Research, Development and Plant Practice, Proc. of the 3rd Int. Symp. on Hydromet., TMS-AIME, Atlanta, March 1983, p391-406. Dry, M . J., Coetzee, C. F. B., Recovery of gold from refractory ores., Gold 100, vol 2 Extractive Metallurgy of Gold, (South African Inst of Mining & Metallurgy Symposium Series S8), Johannesburg, S.A., p 259-274. Eisele, J. A . , Gold metallurgy - a historical perspective., Canadian Metallurgical Quarterly v 27 n 4 0ct-Dec 1988 p 287-291. Everett, P. K. , Development of the INTEC copper process by an industrial consortium., Presented at: Hydrometallurgy 1994 Symposium, IMM/SCI, Cambridge, U.K. , July 11-15, 1994. Foo, K. A. , Bath, M . D., Ismay, A. , Canterford, J. H. , New gold processing technologies. An engineer's perspective., Gold Forum Technol Pract World Gold 89. Publ by Australasian Inst of Mining & Metallurgy, Parkville, Aust. p 233-250. Fraser, K. S., Walton, R. H. , Wells, J. A . , Processing of refractory gold ores., Minerals Engineering, v 4 n 7-11, 1991, p 1029-1041. Fuchs, W. A. , Schurer, V . C , A mineralogical and chemical analysis of carbonaceous ore at the Santa Fe Mine, Mineral County, Nevada: The K 2 0 - preg rob relation., EPD Congress '90, D.R. Gaskell ed., The Minerals, Metals & Materials Society, p 521-527. 82 Gabra, G., A kinetic study of the leaching of gold from pyrite concentrate using acidified thiourea., Precious Metals, Mining, Extraction and Processing, A I M E Annual Meet., Los Angeles, 1984,p 145-172. Goode, J. R., Refractory gold ore processing., EPD Congress '93, J. P. Hager ed., The Minerals, Metals & Materials Society, p 291-317. Graham, J., A mineralogical procedure for recognising cyanicides in gold processing., Fifth AusIMM Extractive Metallurgy Conference. Publ by Australasian Inst of Mining & Metallurgy, Parkville, Aust., Oct. 1991, p 69-73. Greaves, J. N . , Palmer, G. R., White, W. W., Recovery of gold from refractory ores by the use of carbon-in-chlorine leaching., JOM v 42 n 9, Sep 1990, p 12-14. Green, B. R., Schwellnus, A . H. , Kotze, M . H. , Recent developments in resins for the extraction of gold., Gold 100, vol 2 Extractive Metallurgy of Gold, (South African Inst of Mining & Metallurgy Symposium Series S8), Johannesburg, S.A., p 321-333. Grimsey, E. J., Aylmore, M . G., Roasting of arsenopyrite., Proc of the Gold 90 Symposium, Publ by Soc of Mining Engineers of AIME, ,Littleton, CO, USA. p 397-410. Guay, W. J., The treatment of refractory gold ores containing carbonaceous material and sulphides., Gold and Silver Leaching, Recovery and Economics, Proc. 110th A I M E Meeting, Chicago, 1981, p 17-22. Guay, W. J., Recovery of gold from carbonaceous gold-bearing ores., U.S. Patent 4,188,208, Dec. 22,1978. Habashi, F., Kinetics and mechanism of gold and silver dissolution in cyanide solution., Bulletin 59 [1967], The Encyclopedia of Gold and Silver Leaching, R.A. Hatkoff ed.,Inter Resources Ltd., Denver, Colorado, 1983, pp 42. Hackl, Ralph P., Operation of a commercial scale bioleach reactor at the Congress gold property., Advances in Gold and Silver Processing Proc Symp GOLDTech 4. Publ by Soc for Mining, Metallurgy & Exploration Inc, Littleton, CO, US A . p 131 -141. Hagni, A . M . , Hagni, R. D.,Taylor, P. R., Mineralogical and textural characterization of lime roasted pyrite and arsenopyrite for gold leaching., Process Mineralogy XII, The Minerals, Metals & Materials Society, 420 Commonwealth Dr., Warrendale, Pennsylvania, 1994, p 141-149. Haque, K. E., The role of oxygen in cyanide leaching of gold ore., C I M Bulletin, September 1992, p 31-37. 83 Harris, L. , Brierley, J. A . , Biotechnology., Mining Magazine v 161 n 4 Oct 1989 p 301-302, 304 Hausen, D. M . , Reversible Reactions Between Pyrite and Pyrrhotite in S0 2 , JOM 43, (4), Apr. 1991, p 31-34. Hausen, D. M . , Processing gold quarry refractory ores., JOM v 41 n 4 Apr 1989 p 43-45. Hefter, G.T., May, P.M. , Chemical speciation in hydrometallurgical cyanide solutions., Fifth AusIMM Extractive Metallurgy Conference. Publ by Australasian Inst of Mining & Metallurgy, Parkville, Aust., Oct. 1991, p 139-146. Heinen, H . J., Eisele, J. A. , Scheiner, B. J., Malononitrile extraction of gold from ores., U S B M RI 7464, 1970. Hemmati, M . , Hendrix, J. L., Nelson, J. H., Milosavljevic, E. B., Study of the thiosulfate leaching of gold from carbonaceous ore and the quantitative determination of thiosulfate in the leached solution., Extraction Metallurgy '89 Symposium, Inst. Min. Metall, London, p665-678. Henley, K . J., Combined mineralogical/metallurgical approach to determining the nature and location of gold in ores and mill products. Minerals Engineering, v 2 n 4 1989, p 459-470. Hiskey, J. B. , Atluri, V . P., Dissolution chemistry of gold and silver in different lixiviants., Min. Proc. and Extr. Met. Review, vol 4, 1988, p 95-134. Hiskey, J. B., Thiourea leaching of gold and silver - technology update and additional applications., Practical Hydromet '83, Annual Symposium on Uranium and Precious Metals 7th. Publ by Soc of Mining Engineers of A I M E , New York, N Y , USA, p 95-99. Hiskey, J. B. , Atluri, V . P., Resin-in-leach cyanidation of a carbonaceous gold ore., Minerals & Metallurgical Processing v 6 n 3, Aug 1989, p 113-116. Hiskey, J. B., Thiourea as a lixiviant for gold and silver., Gold and Silver Leaching, Recovery and Economics, Proc. 110th A I M E Meeting, Chicago, 1981, p 83-92. Hsu, T.L., Tran, T., Young, D., Modelling of the chemical species-application to effluent treatment., Fifth AusIMM Extractive Metallurgy Conference. Publ by Australasian Inst of Mining & Metallurgy, Parkville, Aust., Oct. 1991, p 133-138. Ibrado, A.S. , Fuerstenau, D.W., Effect of the structure of carbon adsorbents on the adsorption of gold cyanide., Hydrometallurgy Theory and Practice, Hydrometallurgy, 30 (1992), p 243-256. 84 Johnson, G., Corrans, I., Angove, J., The Activox™ process for refractory gold ores., Preprint, Randol Gold Forum '93, Beaver Creek, Colorado, 1993. Karavaiko, G. I., Chuchalin, L. K. , Pivovarova, T. A . , Yemel'yanov, B. A . , Dorofeyev, A . G., Microbiological Leaching of Metals From Arsenopyrite Containing Concentrates., Fundamental and Applied Biohydrometallurgy, Elsevier Science Publishers, P.O. Box 211, 1000 A E Amsterdam, The Netherlands, 1986, p 115-126. Kerley, B.J. Jr., Recovery of precious metals from difficult ores., U.S. Patent 4,269,622, May 26, 1981. Kerley, B.J. Jr., Recovery of precious metals from difficult ores., U.S. Patent 4,369,061, January 18, 1983. Komnitsas, C , Pooley, F. D., Bacterial oxidation of an arsenical gold sulphide concentrate from Olympias, Greece., Minerals Engineering, v 3 n 3-4, 1990, p 295-306. Komnitsas, C , Pooley, F. D., Optimization of the bacterial oxidation of an arsenical gold sulphide concentrate from Olympias, Greece., Minerals Engineering, v 4 n 12 1991, p 1297-1303. Komnitsas, C , Pooley, F. D., Mineralogical characteristics and treatment of refractory gold ores., Minerals Engineering, v 2 n 4, 1989, p 449-457. Kondos, P. D., Deschenes, G., Morrison, R. M . , Process optimization studies in gold cyanidation., Hydrometallurgy 39, 1995, p 235-250. Lakshmanan, V . I., Bosch, D. W., Child, A . M . , Treatment of refractory gold ores. Leaching, Unit Operations and Processes, 14th Annual CIM Hydrometallurgical Meeting, Timmins, Ontario, October 1984. Langhans, J. W. Jr., Lei, K. P. V. , Carnahan, T. G., Copper catalyzed thiosulphate leaching of low-grade gold ores., Hydrometallurgy Theory and Practice, Proceedings of the Ernest Peters International Symposium. Part A. Hydrometallurgy 29 (1992), vol 1-3, p 191-204. Lee, A. F., Brown, T. D., Laxen, P. A. , The distribution of humic substances in the dumps and CIP plant circuits at Rand Mines milling and mining., Gold 100, vol 2 Extractive Metallurgy of Gold, (South African Inst of Mining & Metallurgy Symposium Series S8), Johannesburg, S.A., p367-382. Liu, X . , Lindstrom, B. , Petersson, S., Bioleaching of refractory arsenical pyrite concentrates to enhance gold extraction., Scandinavian Journal of Metallurgy v 20 n 6, 1991, p 346-350. 85 Lorenzen, L . , van Deventer, J.S.J., Identification of refractoriness in gold ores by the selective destruction of minerals., Minerals Engineering, v 6 n 8-10, Aug-Oct 1993, p 1013-1023. Major, K . W., Semple, P. G., Nendick, R. M . , Selection and evaluation of a refractory gold extraction process to suit an orebody., Proc of the Gold 90 Symposium, Publ by Soc of Mining Engineers of AIME, ,Littleton, CO, USA., p 275-280. Mason, P. G., Energy requirements for the pressure oxidation of gold bearing sulfides., JOM, Sept 1990, p 15-18. McGrew, K . J., Murphy, J. W., Iodine leach for the dissolution of gold., U.S. Patent 4,557,759., April 10, 1984. McKay, D. R., Halpern, J., A kinetic study of the oxidation of pyrite in aqueous suspension., Trans. A I M E , June 1958, p 301-309. McLelland, G. E., Wroblenski, M . D., Eisele, J. A. , Improved processing technology for recovering precious metals from low grade ores., Practical Hydromet '83,7th Annual Symp. on Uranium and Precious Metals., SME-AIME, Lakewood, Colo., 1983, p 93-94. Mishra, K. K., Osseo-Asare, K., Aspects of the interfacial electrochemistry of semiconductor pyrite (FeS2)., J. Electrochem. Soc, vol 135 (1988), p 2502-2509. Mrkusic, P.D.P., The recovery of gold from sulphide and arsenical ores mainly from the Barberton area., N I M , Randburg, S.Africa, Report No. 911, Aug. 1970. Muir, C. W. A. , Hendriks, L . P., Gussman, H. W., The treatment of refractory gold bearing concentrates using pressure leaching techniques., Precious Metals: Mining, Extraction and Processing, Proceeds of an International Symposium by TMS of A I M E and IPMI, A I M E Annual Meeting, Los Angeles, Calif, February 1984. Nagpal, S., Oolman, T., Free, M . L. , Palmer, B., Dahlstrom, D., Biooxidation of a Refractory Pyrite-Arsenopyrite-Gold Ore Concentrate., Mineral Bioprocessing, The Minerals, Metals & Materials Society, 420 Commonwealth Dr., Warrendale, Pennsylvania 1991, 469-483 Nagy, I., Mrkusic, P., McCulloch, H. W., Chemical treatment of refractory gold ores. A literature survey., NIM, Johannesburg, S. Africa, Report No. 38, June 1966. Nicol, M . J., The non-oxidative leaching of oxides and sulphides : An electrochemical approach., Hydrometallurgy Research, Development and Plant Practice, Proc. of the 3rd Int. Symp. on Hydromet., TMS-AIME, Atlanta, March 1983, p 177-196. 86 Osseo-Asare, K. , Semiconductor electrochemistry and hydrometallurgical dissolution processes., Hydrometallurgy, 29, 1992, p 61-90. Pantelis, G . , Ritchie, A . I. M . , Macroscopic Transport Mechanisms as a Rate-Limiting Factor in Dump Leaching of Pyritic Ores., Applied Mathematical Modelling 15, (3), Mar. 1991, p 136-143. Papangelakis, V . G , Demopoulos, G. P., Acid pressure oxidation of arsenopyrite. II. Reaction kinetics., Can. Metall. Q. 29, (1), p 13-20 Jan.-Mar. 1990. Papangelakis, V . G., Demopoulos, G. P., Acid pressure oxidation of pyrite: reaction kinetics., Hydrometallurgy 26,(3), p 309-325 May 1991. Papangelakis, V . G., Demopoulos, G. P., Acid pressure oxidation of arsenopyrite. I. Reaction chemistry., Can. Metall. Q. 29, (1), Jan.-Mar. 1990, p 1-12 . Pendreigh, R , Barratt, D. J., Refractory gold process options., XVIII Convencion de Ingenieros de Minas, Lima Peru, November 1986. Perez, A . E., Galaviz, H . D., Method for recovery of precious metals from difficult ores with copper-ammonium thiosulfate., U.S. Patent 4,654,078, March 31, 1987. Pesic, B., Sergent, R. H., The rotating disc study of gold dissolution with Geobrom™ 3400., EPD Congress '90, D.R. Gaskell ed., The Minerals, Metals & Materials Society, p 355-391. Pesic, B. , Smith, B . D., Sergent, R. H , Dissolution of gold with bromine from refractory ores preoxidized by pressure oxidation., EPD Congress 1992, The Minerals, Metals & Materials Society, 420 Commonwealth Dr., Warrendale, Pennsylvania, 1991, p 223-237. Potter, G. M . , Salisbury, H . P., Innovations in gold metallurgy., Mining Congress Journal, July 1974. Potter, George M . , Precious metal processing., Annual Symposium on Uranium and Precious Metals 7th. Publ by Soc of Mining Engineers of AIME, New York, N Y , USA, p 89-90. Ralph, D., Bacterial cultures and bio-oxidation., Fifth AusIMM Extractive Metallurgy Conference. Publ by Australasian Inst of Mining & Metallurgy, Parkville, Aust., Oct. 1991, p 147-151. Raudsepp, R., Algood, R., Thiourea leaching of gold in a continuous pilot plant., Proc. of the Int. Symp. on Gold Metallurgy, CIM, Winnipeg, 1987, p 87-96. Scheiner, B. J., Lindstrom, R. E., Henrie, T. A. , Oxidation process for improving gold recovery from carbon-bearing gold ores., U S B M RI 7573,14pp., 1972. 87 Scheiner, B. J., Lindstrom, R. E., Henrie, T. A. , Investigation of oxidation systems for improving gold recovery from carbonaceous materials.,USBM TPR2, 8pp., 1968. Sergent, R. H. , The use of brominated hydantoins for gold leaching., Randol Gold Forum 1988, pl67-169. Sibrell, P. L . , Wan, R. Y . , Miller, J. D., Spectroscopic analysis of passivation reactions for carbonaceous matter from Carlin trend ores., Proc of the Gold 90 Symposium, Publ by Soc of Mining Engineers of AIME, ,Littleton, CO, USA. p 355-363. Spencer, P. A. , Budden, J. R., Metallurgical considerations in the design of a bio-oxidation plant: A case study for a refractory arsenopyrite concentrate., EPD Congress '90, D.R. Gaskell ed., The Minerals, Metals & Materials Society, p 295-303. Stanley, G.G., ed., The Extractive Metallurgy of Gold in South Africa, S A I M M , Johannesburg, 1987, vol. 2. Stoychevski, M . , Williams, L . R., Influence of oxygen, hydrogen peroxide and Oxone on dissolution of gold from pyrite ore., Trans. Instn Min. Metall. C, 102, May-August 1993, pC93-C98. Swash, P. M . , A mineralogical investigation of refractory gold ores and their beneficiation, with special reference to arsenical ores., Journal of The South African Institute of Mining and Metallurgy v 88 n 5 May 1988 p 173-180. Swash, P. M . , A mineralogical investigation of refractory gold ores and their beneficiation., MINTEK Report No. 295, December 1986, Pp 15. Swash, P. M . , Ellis, P., The roasting of arsenical gold ores: A mineralogical perspective., Gold 100, Proceeding of the International Conference on Gold, vol 2, p 235-257. Torma, A . E., New trends in biohydrometallurgy., Mineral Bioprocessing (proceedings), R. W. Smith and M . Misra eds., The Minerals, Metals & Materials Society, 1991, p 43-55. Touro, F. J., Wiewiorowski, T. K., Process for the recovery of gold from refractory gold ores using sulfurous acid as the leaching agent., U.S. Patent #5,147,618, September 15, 1992. Tran, T., Davis, A . , Fundamental aspects of the leaching of gold in halide media., EPD Congress '92, J. P. Hager ed., The Minerals, Metals & Materials Society, p 99-113. Tran, T., Davis, A. , Gold dissolution in iodide electrolytes., Hydrometallurgy, 26,1991, p 163-177. 88 Tumilty J. A. , Sweeney, A. G., Lorenzen, L., Diagnostic leaching in the development of flowsheets for new ore deposits., Proc. of the Int. Symp. on Gold Metallurgy, C I M , Winnipeg, 1987, p 157-168. van Aswegen, P. C , Godfrey, M . W., Miller, D. M . , Haines, A. K. , Developments and innovations in bacterial oxidation of refractory ores., Minerals & Metallurgical Processing v 8 n 4 Nov 1991, p 188-191. van Weert, G., Fair, K . J., Schneider, J. C , Prochem's Nitrox process., 88th Annual Gen Meet., CIM, Montreal, Quebec, May 1986. Vaughan, J. P., Mineralogical factors in the processing of Archean sulphide gold ores., Fifth AusIMM Extractive Metallurgy Conference. Publ by Australasian Inst of Mining & Metallurgy, Parkville, Aust., Oct. 1991, p 61-67. Yen, W. T., Wyslouzil, D. M . , Pressure oxidation and thiourea extraction of refractory gold ore., Gold 100, vol 2 Extractive Metallurgy of Gold, (South African Inst of Mining & Metallurgy Symposium Series S8), Johannesburg, S.A., p 579-589. Yen, W. T., Pindred, R. A. , Lam, M . P., Hypochlorite leaching of gold ore., Advances in Gold and Silver Processing Proc Symp GOLDTech 4. Publ by Soc for Mining, Metallurgy & Exploration Inc, Littleton, CO, USA, p 67-74. Yen, W. T., Pindred, R. A. , Pressure oxidation of refractory gold ore with sodium hypochlorite., Precious Metals 1989, Proc. of the 13th IPMI Conf, Montreal, Quebec, June 1989, p 335-343. Zipperian, D., Raghavan, S., Wilson, J. P., Gold and silver extraction by ammoniacal thiosulfate leaching from a rhyolite ore., Hydrometallurgy, 19,1988, p 361-375. 89 A P P E N D I X 90 0 X 'X a m o +> 0 3) C o i i£ IS iJ) u ) « o J03 - I N C O " D— «JCDG0 - < E Z H C 0) s; x i—< < x iK0r>£r> r i c r i ' j ' r - c o v j j i Q Q Q Q I S I I I I I 1 • ic-tcnvDrOuD cn ro 11 "1 LD 0 T - :( 0 : 1 s y © « © L 0 ^ 0 0 ' 0 £ 8 s ' + u n o Q : T LTJ O -C '' • 0 C? w i e-*>«4z:a;c9 ' B C. 3 rH « O - N I O — C — ^ o © -* *t U i n o - • lO o J l 0 f'l X CD O JlCD -< CD n ••••• M X ID-t> CO u.iO+ ! a> •>•' O "w u CO 80 CO <— •Q - * c . o c o c o s ; I Z >i> C i) cn SODI^^EFir-ID •;cn r^[--CDsrfsi I I I I f 1 I , v< cr» ..£1 m •> ro 1 ro ro en H to ^ 0 1 * 8 9 1 0 0 ' 0 1* • +> m (S X 0 3 (X CO CN m 6 sq.unoQ 0 0 - 0 0 Q in •J 3 J) •Ti 13 C. - J) Q t ) •JD O C3 M i *^ —• : in n3 1 «H 0.1 w w 0 •4 (D+> 3 C') C « 0 c c oo • i 1-4 m *> o —* £, CO •>oH - -TO Of - «N 0 — U C O 0 3'lCO U W-CO X i o o •-L0X 3 0 ' • - o CM CO ft C0 03 rj CO m ri)